fraud_detection_example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fraud Detection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* [Reading Data](#1)\n",
    "* [Histograms](#2)\n",
    "* [Correlations](#3)\n",
    "* [WoE](#4)\n",
    "* [Outliers](#5)\n",
    "* [Features with Low Variance](#6)\n",
    "* [PCA](#7)\n",
    "* [Modeling](#8)\n",
    "* [GridSearch](#9)\n",
    "* [Conclusion](#10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy.stats import zscore, ks_2samp\n",
    "from sklearn.preprocessing import MinMaxScaler, StandardScaler\n",
    "from sklearn.feature_selection import VarianceThreshold\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import confusion_matrix, classification_report, accuracy_score, roc_auc_score\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import LinearSVC, SVC\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from collections import Counter\n",
    "\n",
    "pd.set_option('chained_assignment',None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reading Data <a class=\"anchor\" id=\"1\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150000, 31)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ocorrencia</th>\n",
       "      <th>PP1</th>\n",
       "      <th>PP2</th>\n",
       "      <th>PP3</th>\n",
       "      <th>PP4</th>\n",
       "      <th>PP5</th>\n",
       "      <th>PP6</th>\n",
       "      <th>PP7</th>\n",
       "      <th>PP8</th>\n",
       "      <th>PP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PP21</th>\n",
       "      <th>PP22</th>\n",
       "      <th>PP23</th>\n",
       "      <th>PP24</th>\n",
       "      <th>PP25</th>\n",
       "      <th>PP26</th>\n",
       "      <th>PP27</th>\n",
       "      <th>PP28</th>\n",
       "      <th>Sacado</th>\n",
       "      <th>Fraude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-44299.0</td>\n",
       "      <td>-1.239996</td>\n",
       "      <td>0.985194</td>\n",
       "      <td>-1.005080</td>\n",
       "      <td>0.251323</td>\n",
       "      <td>0.872854</td>\n",
       "      <td>-1.677811</td>\n",
       "      <td>1.451311</td>\n",
       "      <td>-0.478908</td>\n",
       "      <td>-0.009459</td>\n",
       "      <td>...</td>\n",
       "      <td>0.387768</td>\n",
       "      <td>0.286200</td>\n",
       "      <td>0.128686</td>\n",
       "      <td>1.280392</td>\n",
       "      <td>-0.301116</td>\n",
       "      <td>-0.673309</td>\n",
       "      <td>-0.069611</td>\n",
       "      <td>-0.009597</td>\n",
       "      <td>-28.38</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-44300.0</td>\n",
       "      <td>-0.472690</td>\n",
       "      <td>1.869177</td>\n",
       "      <td>-0.277741</td>\n",
       "      <td>1.122846</td>\n",
       "      <td>1.526166</td>\n",
       "      <td>0.262325</td>\n",
       "      <td>0.242333</td>\n",
       "      <td>-0.006108</td>\n",
       "      <td>-1.659659</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.387745</td>\n",
       "      <td>-0.434629</td>\n",
       "      <td>0.512801</td>\n",
       "      <td>-0.110994</td>\n",
       "      <td>-0.350975</td>\n",
       "      <td>-0.073826</td>\n",
       "      <td>0.035071</td>\n",
       "      <td>-0.080140</td>\n",
       "      <td>-407.00</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-44301.0</td>\n",
       "      <td>0.277314</td>\n",
       "      <td>3.455314</td>\n",
       "      <td>-0.722444</td>\n",
       "      <td>-0.428284</td>\n",
       "      <td>2.512025</td>\n",
       "      <td>-0.540760</td>\n",
       "      <td>0.345111</td>\n",
       "      <td>-0.013655</td>\n",
       "      <td>-0.233508</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.630255</td>\n",
       "      <td>-0.388096</td>\n",
       "      <td>0.697177</td>\n",
       "      <td>-0.523084</td>\n",
       "      <td>-0.069830</td>\n",
       "      <td>0.196482</td>\n",
       "      <td>0.052145</td>\n",
       "      <td>-0.166683</td>\n",
       "      <td>-800.00</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-44301.0</td>\n",
       "      <td>-1.061770</td>\n",
       "      <td>-0.105481</td>\n",
       "      <td>-0.226711</td>\n",
       "      <td>-0.929524</td>\n",
       "      <td>-0.100625</td>\n",
       "      <td>-0.300173</td>\n",
       "      <td>0.029912</td>\n",
       "      <td>-0.205934</td>\n",
       "      <td>0.233190</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.147422</td>\n",
       "      <td>-0.426827</td>\n",
       "      <td>0.070413</td>\n",
       "      <td>0.283090</td>\n",
       "      <td>-0.487739</td>\n",
       "      <td>0.288220</td>\n",
       "      <td>-0.035644</td>\n",
       "      <td>-0.007305</td>\n",
       "      <td>-31.28</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-44302.0</td>\n",
       "      <td>4.622715</td>\n",
       "      <td>2.621667</td>\n",
       "      <td>0.872085</td>\n",
       "      <td>0.374010</td>\n",
       "      <td>1.456021</td>\n",
       "      <td>-1.531875</td>\n",
       "      <td>-0.162837</td>\n",
       "      <td>-1.331547</td>\n",
       "      <td>-0.340639</td>\n",
       "      <td>...</td>\n",
       "      <td>0.221196</td>\n",
       "      <td>0.804017</td>\n",
       "      <td>1.309062</td>\n",
       "      <td>1.505088</td>\n",
       "      <td>0.260178</td>\n",
       "      <td>-0.861611</td>\n",
       "      <td>-0.130562</td>\n",
       "      <td>1.023781</td>\n",
       "      <td>-522.16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Ocorrencia       PP1       PP2       PP3       PP4       PP5       PP6  \\\n",
       "0    -44299.0 -1.239996  0.985194 -1.005080  0.251323  0.872854 -1.677811   \n",
       "1    -44300.0 -0.472690  1.869177 -0.277741  1.122846  1.526166  0.262325   \n",
       "2    -44301.0  0.277314  3.455314 -0.722444 -0.428284  2.512025 -0.540760   \n",
       "3    -44301.0 -1.061770 -0.105481 -0.226711 -0.929524 -0.100625 -0.300173   \n",
       "4    -44302.0  4.622715  2.621667  0.872085  0.374010  1.456021 -1.531875   \n",
       "\n",
       "        PP7       PP8       PP9   ...        PP21      PP22      PP23  \\\n",
       "0  1.451311 -0.478908 -0.009459   ...    0.387768  0.286200  0.128686   \n",
       "1  0.242333 -0.006108 -1.659659   ...   -0.387745 -0.434629  0.512801   \n",
       "2  0.345111 -0.013655 -0.233508   ...   -0.630255 -0.388096  0.697177   \n",
       "3  0.029912 -0.205934  0.233190   ...   -0.147422 -0.426827  0.070413   \n",
       "4 -0.162837 -1.331547 -0.340639   ...    0.221196  0.804017  1.309062   \n",
       "\n",
       "       PP24      PP25      PP26      PP27      PP28  Sacado  Fraude  \n",
       "0  1.280392 -0.301116 -0.673309 -0.069611 -0.009597  -28.38       0  \n",
       "1 -0.110994 -0.350975 -0.073826  0.035071 -0.080140 -407.00       0  \n",
       "2 -0.523084 -0.069830  0.196482  0.052145 -0.166683 -800.00       0  \n",
       "3  0.283090 -0.487739  0.288220 -0.035644 -0.007305  -31.28       0  \n",
       "4  1.505088  0.260178 -0.861611 -0.130562  1.023781 -522.16       0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('desafio_fraude.csv')\n",
    "\n",
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Quick Check"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 150000 entries, 0 to 149999\n",
      "Data columns (total 31 columns):\n",
      "Ocorrencia    150000 non-null float64\n",
      "PP1           150000 non-null float64\n",
      "PP2           150000 non-null float64\n",
      "PP3           150000 non-null float64\n",
      "PP4           150000 non-null float64\n",
      "PP5           150000 non-null float64\n",
      "PP6           150000 non-null float64\n",
      "PP7           150000 non-null float64\n",
      "PP8           150000 non-null float64\n",
      "PP9           150000 non-null float64\n",
      "PP10          150000 non-null float64\n",
      "PP11          150000 non-null float64\n",
      "PP12          150000 non-null float64\n",
      "PP13          150000 non-null float64\n",
      "PP14          150000 non-null float64\n",
      "PP15          150000 non-null float64\n",
      "PP16          150000 non-null float64\n",
      "PP17          150000 non-null float64\n",
      "PP18          150000 non-null float64\n",
      "PP19          150000 non-null float64\n",
      "PP20          150000 non-null float64\n",
      "PP21          150000 non-null float64\n",
      "PP22          150000 non-null float64\n",
      "PP23          150000 non-null float64\n",
      "PP24          150000 non-null float64\n",
      "PP25          150000 non-null float64\n",
      "PP26          150000 non-null float64\n",
      "PP27          150000 non-null float64\n",
      "PP28          150000 non-null float64\n",
      "Sacado        150000 non-null float64\n",
      "Fraude        150000 non-null int64\n",
      "dtypes: float64(30), int64(1)\n",
      "memory usage: 35.5 MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are no null-values and all variables are continuous. Just to make sure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('Ocorrencia', 0),\n",
       " ('PP1', 0),\n",
       " ('PP2', 0),\n",
       " ('PP3', 0),\n",
       " ('PP4', 0),\n",
       " ('PP5', 0),\n",
       " ('PP6', 0),\n",
       " ('PP7', 0),\n",
       " ('PP8', 0),\n",
       " ('PP9', 0),\n",
       " ('PP10', 0),\n",
       " ('PP11', 0),\n",
       " ('PP12', 0),\n",
       " ('PP13', 0),\n",
       " ('PP14', 0),\n",
       " ('PP15', 0),\n",
       " ('PP16', 0),\n",
       " ('PP17', 0),\n",
       " ('PP18', 0),\n",
       " ('PP19', 0),\n",
       " ('PP20', 0),\n",
       " ('PP21', 0),\n",
       " ('PP22', 0),\n",
       " ('PP23', 0),\n",
       " ('PP24', 0),\n",
       " ('PP25', 0),\n",
       " ('PP26', 0),\n",
       " ('PP27', 0),\n",
       " ('PP28', 0),\n",
       " ('Sacado', 0),\n",
       " ('Fraude', 0)]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[(col, df[col].isna().sum()) for col in df.columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Really no null-values. One less headache! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ocorrencia</th>\n",
       "      <th>PP1</th>\n",
       "      <th>PP2</th>\n",
       "      <th>PP3</th>\n",
       "      <th>PP4</th>\n",
       "      <th>PP5</th>\n",
       "      <th>PP6</th>\n",
       "      <th>PP7</th>\n",
       "      <th>PP8</th>\n",
       "      <th>PP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PP21</th>\n",
       "      <th>PP22</th>\n",
       "      <th>PP23</th>\n",
       "      <th>PP24</th>\n",
       "      <th>PP25</th>\n",
       "      <th>PP26</th>\n",
       "      <th>PP27</th>\n",
       "      <th>PP28</th>\n",
       "      <th>Sacado</th>\n",
       "      <th>Fraude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "      <td>150000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-84550.214580</td>\n",
       "      <td>0.058999</td>\n",
       "      <td>-0.000790</td>\n",
       "      <td>-0.192183</td>\n",
       "      <td>-0.037416</td>\n",
       "      <td>0.061588</td>\n",
       "      <td>-0.025715</td>\n",
       "      <td>0.026695</td>\n",
       "      <td>-0.004257</td>\n",
       "      <td>0.028148</td>\n",
       "      <td>...</td>\n",
       "      <td>0.009957</td>\n",
       "      <td>0.027398</td>\n",
       "      <td>0.007275</td>\n",
       "      <td>-0.002739</td>\n",
       "      <td>-0.035211</td>\n",
       "      <td>-0.001127</td>\n",
       "      <td>-0.000535</td>\n",
       "      <td>-0.001028</td>\n",
       "      <td>-88.602261</td>\n",
       "      <td>0.001580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>27710.748503</td>\n",
       "      <td>1.894453</td>\n",
       "      <td>1.623712</td>\n",
       "      <td>1.406053</td>\n",
       "      <td>1.397615</td>\n",
       "      <td>1.341265</td>\n",
       "      <td>1.310820</td>\n",
       "      <td>1.194923</td>\n",
       "      <td>1.205874</td>\n",
       "      <td>1.106154</td>\n",
       "      <td>...</td>\n",
       "      <td>0.739429</td>\n",
       "      <td>0.707714</td>\n",
       "      <td>0.622620</td>\n",
       "      <td>0.606964</td>\n",
       "      <td>0.506130</td>\n",
       "      <td>0.483787</td>\n",
       "      <td>0.397662</td>\n",
       "      <td>0.307684</td>\n",
       "      <td>247.302373</td>\n",
       "      <td>0.039718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-133236.000000</td>\n",
       "      <td>-2.454930</td>\n",
       "      <td>-22.057729</td>\n",
       "      <td>-9.382558</td>\n",
       "      <td>-16.875344</td>\n",
       "      <td>-32.911462</td>\n",
       "      <td>-21.307738</td>\n",
       "      <td>-31.527244</td>\n",
       "      <td>-16.635979</td>\n",
       "      <td>-15.594995</td>\n",
       "      <td>...</td>\n",
       "      <td>-27.202839</td>\n",
       "      <td>-10.503090</td>\n",
       "      <td>-19.002942</td>\n",
       "      <td>-4.022866</td>\n",
       "      <td>-7.519589</td>\n",
       "      <td>-3.220178</td>\n",
       "      <td>-12.152401</td>\n",
       "      <td>-22.620072</td>\n",
       "      <td>-19656.530000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-115169.000000</td>\n",
       "      <td>-1.243456</td>\n",
       "      <td>-0.802149</td>\n",
       "      <td>-1.138473</td>\n",
       "      <td>-0.812624</td>\n",
       "      <td>-0.526469</td>\n",
       "      <td>-0.424574</td>\n",
       "      <td>-0.527260</td>\n",
       "      <td>-0.340863</td>\n",
       "      <td>-0.565387</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.165038</td>\n",
       "      <td>-0.466423</td>\n",
       "      <td>-0.128298</td>\n",
       "      <td>-0.431560</td>\n",
       "      <td>-0.369398</td>\n",
       "      <td>-0.247606</td>\n",
       "      <td>-0.090965</td>\n",
       "      <td>-0.078861</td>\n",
       "      <td>-77.662500</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-77502.500000</td>\n",
       "      <td>0.042647</td>\n",
       "      <td>-0.082193</td>\n",
       "      <td>-0.359076</td>\n",
       "      <td>-0.039549</td>\n",
       "      <td>0.124219</td>\n",
       "      <td>0.245177</td>\n",
       "      <td>-0.013129</td>\n",
       "      <td>-0.037083</td>\n",
       "      <td>0.095975</td>\n",
       "      <td>...</td>\n",
       "      <td>0.033794</td>\n",
       "      <td>0.014600</td>\n",
       "      <td>0.020008</td>\n",
       "      <td>-0.049357</td>\n",
       "      <td>-0.071030</td>\n",
       "      <td>0.057265</td>\n",
       "      <td>-0.004792</td>\n",
       "      <td>-0.016759</td>\n",
       "      <td>-22.040000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-61713.750000</td>\n",
       "      <td>0.952018</td>\n",
       "      <td>0.588600</td>\n",
       "      <td>0.555060</td>\n",
       "      <td>0.816575</td>\n",
       "      <td>0.751890</td>\n",
       "      <td>0.734024</td>\n",
       "      <td>0.564334</td>\n",
       "      <td>0.193112</td>\n",
       "      <td>0.678488</td>\n",
       "      <td>...</td>\n",
       "      <td>0.225362</td>\n",
       "      <td>0.540801</td>\n",
       "      <td>0.164620</td>\n",
       "      <td>0.348762</td>\n",
       "      <td>0.274183</td>\n",
       "      <td>0.331361</td>\n",
       "      <td>0.068544</td>\n",
       "      <td>0.048427</td>\n",
       "      <td>-5.410000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-44299.000000</td>\n",
       "      <td>36.802320</td>\n",
       "      <td>63.344698</td>\n",
       "      <td>33.680984</td>\n",
       "      <td>5.683171</td>\n",
       "      <td>31.356750</td>\n",
       "      <td>21.929312</td>\n",
       "      <td>43.557242</td>\n",
       "      <td>73.216718</td>\n",
       "      <td>13.434066</td>\n",
       "      <td>...</td>\n",
       "      <td>34.830382</td>\n",
       "      <td>10.933144</td>\n",
       "      <td>44.807735</td>\n",
       "      <td>2.824849</td>\n",
       "      <td>10.295397</td>\n",
       "      <td>2.604551</td>\n",
       "      <td>22.565679</td>\n",
       "      <td>11.710896</td>\n",
       "      <td>-0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          Ocorrencia            PP1            PP2            PP3  \\\n",
       "count  150000.000000  150000.000000  150000.000000  150000.000000   \n",
       "mean   -84550.214580       0.058999      -0.000790      -0.192183   \n",
       "std     27710.748503       1.894453       1.623712       1.406053   \n",
       "min   -133236.000000      -2.454930     -22.057729      -9.382558   \n",
       "25%   -115169.000000      -1.243456      -0.802149      -1.138473   \n",
       "50%    -77502.500000       0.042647      -0.082193      -0.359076   \n",
       "75%    -61713.750000       0.952018       0.588600       0.555060   \n",
       "max    -44299.000000      36.802320      63.344698      33.680984   \n",
       "\n",
       "                 PP4            PP5            PP6            PP7  \\\n",
       "count  150000.000000  150000.000000  150000.000000  150000.000000   \n",
       "mean       -0.037416       0.061588      -0.025715       0.026695   \n",
       "std         1.397615       1.341265       1.310820       1.194923   \n",
       "min       -16.875344     -32.911462     -21.307738     -31.527244   \n",
       "25%        -0.812624      -0.526469      -0.424574      -0.527260   \n",
       "50%        -0.039549       0.124219       0.245177      -0.013129   \n",
       "75%         0.816575       0.751890       0.734024       0.564334   \n",
       "max         5.683171      31.356750      21.929312      43.557242   \n",
       "\n",
       "                 PP8            PP9      ...                 PP21  \\\n",
       "count  150000.000000  150000.000000      ...        150000.000000   \n",
       "mean       -0.004257       0.028148      ...             0.009957   \n",
       "std         1.205874       1.106154      ...             0.739429   \n",
       "min       -16.635979     -15.594995      ...           -27.202839   \n",
       "25%        -0.340863      -0.565387      ...            -0.165038   \n",
       "50%        -0.037083       0.095975      ...             0.033794   \n",
       "75%         0.193112       0.678488      ...             0.225362   \n",
       "max        73.216718      13.434066      ...            34.830382   \n",
       "\n",
       "                PP22           PP23           PP24           PP25  \\\n",
       "count  150000.000000  150000.000000  150000.000000  150000.000000   \n",
       "mean        0.027398       0.007275      -0.002739      -0.035211   \n",
       "std         0.707714       0.622620       0.606964       0.506130   \n",
       "min       -10.503090     -19.002942      -4.022866      -7.519589   \n",
       "25%        -0.466423      -0.128298      -0.431560      -0.369398   \n",
       "50%         0.014600       0.020008      -0.049357      -0.071030   \n",
       "75%         0.540801       0.164620       0.348762       0.274183   \n",
       "max        10.933144      44.807735       2.824849      10.295397   \n",
       "\n",
       "                PP26           PP27           PP28         Sacado  \\\n",
       "count  150000.000000  150000.000000  150000.000000  150000.000000   \n",
       "mean       -0.001127      -0.000535      -0.001028     -88.602261   \n",
       "std         0.483787       0.397662       0.307684     247.302373   \n",
       "min        -3.220178     -12.152401     -22.620072  -19656.530000   \n",
       "25%        -0.247606      -0.090965      -0.078861     -77.662500   \n",
       "50%         0.057265      -0.004792      -0.016759     -22.040000   \n",
       "75%         0.331361       0.068544       0.048427      -5.410000   \n",
       "max         2.604551      22.565679      11.710896      -0.000000   \n",
       "\n",
       "              Fraude  \n",
       "count  150000.000000  \n",
       "mean        0.001580  \n",
       "std         0.039718  \n",
       "min         0.000000  \n",
       "25%         0.000000  \n",
       "50%         0.000000  \n",
       "75%         0.000000  \n",
       "max         1.000000  \n",
       "\n",
       "[8 rows x 31 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, variables have different ranges and deviations. Maybe will need to feature scale later.\n",
    "\n",
    "Let's look at the target."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.99842\n",
       "1    0.00158\n",
       "Name: Fraude, dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Fraude'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Indeed, it is an **unbalanced** problem, given that only about 0.1% of the records are frauds."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Histograms <a class=\"anchor\" id=\"2\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.iloc[:,0:12].hist(bins=50, figsize=(20,15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.iloc[:,12:24].hist(bins=50, figsize=(20,15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.iloc[:,24:].hist(bins=50, figsize=(20,15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the features are nicely normally distributed, but some seem to have too little variance (like PP7, PP8, PP20, PP21, PP23, PP27 and PP28).  \n",
    "\n",
    "Will just adjust PP1 to better separate the two groups (using Cubic Root transformation)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Adjusting PP1 with Cubic Root"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['PP1'] = np.cbrt(df['PP1'])\n",
    "\n",
    "df['PP1'].hist(bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now the variable bimodal behavior is better separated and outlier effect is diminished."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlations <a class=\"anchor\" id=\"3\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x1800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def correlation_heatmap(train):\n",
    "    correlations = train.corr()\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=(25,25))\n",
    "    sns.heatmap(correlations, vmax=1.0, center=0, fmt='.2f',\n",
    "                square=True, linewidths=.5, annot=True, cbar_kws={\"shrink\": .70})\n",
    "    plt.show()\n",
    "    \n",
    "correlation_heatmap(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the features have insignificant correlations. Most notable are PP3/Ocorrencia and Sacado with other 4 varibles, the highest being -0.55 with PP2 (although this value does not imply very high correlation). Given that, I have decided to not exclude any variable based on correlation. PP17, PP14, PP12 and PP10 have the highest correlation with the target variable. Keep those features in mind."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### WoE <a class=\"anchor\" id=\"4\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To conduct WoE (Weight Of Evidence), one must first categorize numerical variables. Will do that binning every variable in 3 buckets with equal number of observations, using pandas *qcut* function. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ocorrencia</th>\n",
       "      <th>PP1</th>\n",
       "      <th>PP2</th>\n",
       "      <th>PP3</th>\n",
       "      <th>PP4</th>\n",
       "      <th>PP5</th>\n",
       "      <th>PP6</th>\n",
       "      <th>PP7</th>\n",
       "      <th>PP8</th>\n",
       "      <th>PP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PP21</th>\n",
       "      <th>PP22</th>\n",
       "      <th>PP23</th>\n",
       "      <th>PP24</th>\n",
       "      <th>PP25</th>\n",
       "      <th>PP26</th>\n",
       "      <th>PP27</th>\n",
       "      <th>PP28</th>\n",
       "      <th>Sacado</th>\n",
       "      <th>Fraude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Ocorrencia  PP1  PP2  PP3  PP4  PP5  PP6  PP7  PP8  PP9   ...    PP21  \\\n",
       "0           2    0    2    0    1    2    0    2    0    1   ...       2   \n",
       "1           2    1    2    1    2    2    1    1    1    0   ...       0   \n",
       "2           2    1    2    1    1    2    0    1    1    1   ...       0   \n",
       "3           2    1    1    1    0    1    0    1    1    1   ...       0   \n",
       "4           2    2    2    2    1    2    0    1    0    0   ...       2   \n",
       "\n",
       "   PP22  PP23  PP24  PP25  PP26  PP27  PP28  Sacado  Fraude  \n",
       "0     1     2     2     0     0     0     1       1       0  \n",
       "1     0     2     1     0     1     1     0       0       0  \n",
       "2     0     2     0     1     1     2     0       0       0  \n",
       "3     0     1     2     0     2     1     1       1       0  \n",
       "4     2     2     2     2     0     0     2       0       0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = [col for col in df if col != 'Fraude']\n",
    "\n",
    "df_woe = df.copy()\n",
    "for col in df_woe[features].columns:\n",
    "    df_woe[col] = pd.qcut(df_woe[col], 3, labels=False)\n",
    "    \n",
    "df_woe.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Defining WoE function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def woe_total(df, feature, target, plot=True):\n",
    "    '''\n",
    "    Calculate WoE (Weight of Evidence) for given feature\n",
    "    considering all dataset \n",
    "    \n",
    "    Pass dataframe and categorized feature\n",
    "    Return WoE dataframe if plot=False, \n",
    "    otherwise return plot\n",
    "    '''\n",
    "    woe_df = df[[feature, target]]\n",
    "    \n",
    "    totals = woe_df.groupby([feature]).count().reset_index()\n",
    "    \n",
    "    totals['Events'] = woe_df.groupby([feature]).sum().reset_index()[target]\n",
    "    totals['Non-Events'] = totals[target] - totals['Events'] \n",
    "\n",
    "    totals['%Events'] = totals['Events']/(totals['Events']+totals['Non-Events'])\n",
    "    totals['%Non-Events'] = totals['Non-Events']/(totals['Events']+\n",
    "                                                                      totals['Non-Events'])\n",
    "    totals['WoE'] = np.log(totals['%Events']/totals['%Non-Events'])\n",
    "\n",
    "    if plot:\n",
    "        categories = np.unique(woe_df.iloc[:,0])\n",
    "        fig, ax = plt.subplots(figsize=(8,4))\n",
    "\n",
    "        for i, cat in enumerate(categories):\n",
    "            markers = ['o', 'v', 's', '*', 'D', 'P', 'X', '^']\n",
    "            ax.scatter(i, totals.loc[totals[feature] == cat, 'WoE'], label=cat, s=200, marker=markers[i])\n",
    "            ax.legend()\n",
    "        ax.set_xticklabels([''])\n",
    "        ax.axhline(0, color='red')\n",
    "        ax.set_title('WoE Total - {}'.format(feature))\n",
    "        ax.set_ylabel('WoE')\n",
    "    else:\n",
    "        return totals "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PP2</th>\n",
       "      <th>Fraude</th>\n",
       "      <th>Events</th>\n",
       "      <th>Non-Events</th>\n",
       "      <th>%Events</th>\n",
       "      <th>%Non-Events</th>\n",
       "      <th>WoE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>50000</td>\n",
       "      <td>197</td>\n",
       "      <td>49803</td>\n",
       "      <td>0.00394</td>\n",
       "      <td>0.99606</td>\n",
       "      <td>-5.532627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>50007</td>\n",
       "      <td>16</td>\n",
       "      <td>49991</td>\n",
       "      <td>0.00032</td>\n",
       "      <td>0.99968</td>\n",
       "      <td>-8.047010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>49993</td>\n",
       "      <td>24</td>\n",
       "      <td>49969</td>\n",
       "      <td>0.00048</td>\n",
       "      <td>0.99952</td>\n",
       "      <td>-7.641104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PP2  Fraude  Events  Non-Events  %Events  %Non-Events       WoE\n",
       "0    0   50000     197       49803  0.00394      0.99606 -5.532627\n",
       "1    1   50007      16       49991  0.00032      0.99968 -8.047010\n",
       "2    2   49993      24       49969  0.00048      0.99952 -7.641104"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "woe_total(df_woe, 'PP2', 'Fraude', False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this variable, we can see that the first bucket has 10 times more Frauds than the other 2 buckets. This type of analysis can be useful when one has knowledge about variables. Since the dataframe given was anonymized, I will not go further with WoE."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Outliers <a class=\"anchor\" id=\"5\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since most of the variables are normally distributed, a good way to detect Outliers is to use the Z-Score. Let's compute them and put it in a separate dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ocorrencia</th>\n",
       "      <th>PP1</th>\n",
       "      <th>PP2</th>\n",
       "      <th>PP3</th>\n",
       "      <th>PP4</th>\n",
       "      <th>PP5</th>\n",
       "      <th>PP6</th>\n",
       "      <th>PP7</th>\n",
       "      <th>PP8</th>\n",
       "      <th>PP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PP21</th>\n",
       "      <th>PP22</th>\n",
       "      <th>PP23</th>\n",
       "      <th>PP24</th>\n",
       "      <th>PP25</th>\n",
       "      <th>PP26</th>\n",
       "      <th>PP27</th>\n",
       "      <th>PP28</th>\n",
       "      <th>Sacado</th>\n",
       "      <th>Fraude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.452554</td>\n",
       "      <td>-0.985401</td>\n",
       "      <td>0.607243</td>\n",
       "      <td>-0.578142</td>\n",
       "      <td>0.206595</td>\n",
       "      <td>0.604854</td>\n",
       "      <td>-1.260358</td>\n",
       "      <td>1.192228</td>\n",
       "      <td>-0.393617</td>\n",
       "      <td>-0.033998</td>\n",
       "      <td>...</td>\n",
       "      <td>0.510952</td>\n",
       "      <td>0.365689</td>\n",
       "      <td>0.195001</td>\n",
       "      <td>2.114023</td>\n",
       "      <td>-0.525371</td>\n",
       "      <td>-1.389421</td>\n",
       "      <td>-0.173706</td>\n",
       "      <td>-0.027852</td>\n",
       "      <td>0.243518</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.452518</td>\n",
       "      <td>-0.711537</td>\n",
       "      <td>1.151666</td>\n",
       "      <td>-0.060850</td>\n",
       "      <td>0.830176</td>\n",
       "      <td>1.091942</td>\n",
       "      <td>0.219741</td>\n",
       "      <td>0.180462</td>\n",
       "      <td>-0.001535</td>\n",
       "      <td>-1.525840</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.537851</td>\n",
       "      <td>-0.652846</td>\n",
       "      <td>0.811936</td>\n",
       "      <td>-0.178356</td>\n",
       "      <td>-0.623881</td>\n",
       "      <td>-0.150271</td>\n",
       "      <td>0.089540</td>\n",
       "      <td>-0.257123</td>\n",
       "      <td>-1.287488</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.452481</td>\n",
       "      <td>0.615415</td>\n",
       "      <td>2.128527</td>\n",
       "      <td>-0.377129</td>\n",
       "      <td>-0.279669</td>\n",
       "      <td>1.826966</td>\n",
       "      <td>-0.392920</td>\n",
       "      <td>0.266475</td>\n",
       "      <td>-0.007794</td>\n",
       "      <td>-0.236547</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.865822</td>\n",
       "      <td>-0.587094</td>\n",
       "      <td>1.108066</td>\n",
       "      <td>-0.857295</td>\n",
       "      <td>-0.068399</td>\n",
       "      <td>0.408465</td>\n",
       "      <td>0.132474</td>\n",
       "      <td>-0.538396</td>\n",
       "      <td>-2.876641</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.452481</td>\n",
       "      <td>-0.935186</td>\n",
       "      <td>-0.064476</td>\n",
       "      <td>-0.024557</td>\n",
       "      <td>-0.638310</td>\n",
       "      <td>-0.120940</td>\n",
       "      <td>-0.209379</td>\n",
       "      <td>0.002692</td>\n",
       "      <td>-0.167246</td>\n",
       "      <td>0.185366</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.212839</td>\n",
       "      <td>-0.641821</td>\n",
       "      <td>0.101408</td>\n",
       "      <td>0.470917</td>\n",
       "      <td>-0.894097</td>\n",
       "      <td>0.598090</td>\n",
       "      <td>-0.088287</td>\n",
       "      <td>-0.020401</td>\n",
       "      <td>0.231791</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>1.555368</td>\n",
       "      <td>1.615105</td>\n",
       "      <td>0.756921</td>\n",
       "      <td>0.294379</td>\n",
       "      <td>1.039644</td>\n",
       "      <td>-1.149025</td>\n",
       "      <td>-0.158615</td>\n",
       "      <td>-1.100691</td>\n",
       "      <td>-0.333398</td>\n",
       "      <td>...</td>\n",
       "      <td>0.285680</td>\n",
       "      <td>1.097367</td>\n",
       "      <td>2.090826</td>\n",
       "      <td>2.484221</td>\n",
       "      <td>0.583625</td>\n",
       "      <td>-1.778647</td>\n",
       "      <td>-0.326980</td>\n",
       "      <td>3.330732</td>\n",
       "      <td>-1.753154</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>-0.981881</td>\n",
       "      <td>-0.034877</td>\n",
       "      <td>-0.322386</td>\n",
       "      <td>-0.838697</td>\n",
       "      <td>0.332925</td>\n",
       "      <td>0.196228</td>\n",
       "      <td>0.192049</td>\n",
       "      <td>-0.057406</td>\n",
       "      <td>-0.753223</td>\n",
       "      <td>...</td>\n",
       "      <td>0.516861</td>\n",
       "      <td>1.400395</td>\n",
       "      <td>-0.121911</td>\n",
       "      <td>0.317786</td>\n",
       "      <td>-0.708302</td>\n",
       "      <td>1.250213</td>\n",
       "      <td>-0.097305</td>\n",
       "      <td>-0.074044</td>\n",
       "      <td>0.349097</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>-0.971023</td>\n",
       "      <td>-0.174094</td>\n",
       "      <td>0.022727</td>\n",
       "      <td>-0.596579</td>\n",
       "      <td>-0.163207</td>\n",
       "      <td>0.004658</td>\n",
       "      <td>-0.057781</td>\n",
       "      <td>-0.047207</td>\n",
       "      <td>0.239018</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.129952</td>\n",
       "      <td>-0.481487</td>\n",
       "      <td>0.185340</td>\n",
       "      <td>0.481784</td>\n",
       "      <td>-1.189520</td>\n",
       "      <td>0.614738</td>\n",
       "      <td>-0.069092</td>\n",
       "      <td>-0.007839</td>\n",
       "      <td>0.352211</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>1.422906</td>\n",
       "      <td>2.943124</td>\n",
       "      <td>-0.502896</td>\n",
       "      <td>0.271327</td>\n",
       "      <td>-2.055703</td>\n",
       "      <td>0.641306</td>\n",
       "      <td>1.064781</td>\n",
       "      <td>-0.404160</td>\n",
       "      <td>1.005183</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.333678</td>\n",
       "      <td>0.649857</td>\n",
       "      <td>-2.266220</td>\n",
       "      <td>2.286027</td>\n",
       "      <td>0.180772</td>\n",
       "      <td>-0.699595</td>\n",
       "      <td>0.306683</td>\n",
       "      <td>-0.735704</td>\n",
       "      <td>-1.198487</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>-0.994214</td>\n",
       "      <td>0.265920</td>\n",
       "      <td>-0.129913</td>\n",
       "      <td>1.049748</td>\n",
       "      <td>0.553487</td>\n",
       "      <td>0.475634</td>\n",
       "      <td>0.293952</td>\n",
       "      <td>-0.014615</td>\n",
       "      <td>-1.404025</td>\n",
       "      <td>...</td>\n",
       "      <td>0.426987</td>\n",
       "      <td>0.975534</td>\n",
       "      <td>-0.086998</td>\n",
       "      <td>-0.036538</td>\n",
       "      <td>-0.642520</td>\n",
       "      <td>0.584208</td>\n",
       "      <td>-0.051309</td>\n",
       "      <td>-0.015922</td>\n",
       "      <td>0.353666</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.452409</td>\n",
       "      <td>0.893312</td>\n",
       "      <td>0.262019</td>\n",
       "      <td>-0.938725</td>\n",
       "      <td>1.166103</td>\n",
       "      <td>0.781218</td>\n",
       "      <td>0.405424</td>\n",
       "      <td>-0.717286</td>\n",
       "      <td>0.049532</td>\n",
       "      <td>1.476413</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.036519</td>\n",
       "      <td>-0.132637</td>\n",
       "      <td>-0.447360</td>\n",
       "      <td>-0.906489</td>\n",
       "      <td>0.154585</td>\n",
       "      <td>-2.006428</td>\n",
       "      <td>0.161230</td>\n",
       "      <td>-0.315946</td>\n",
       "      <td>-0.464282</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.452373</td>\n",
       "      <td>0.909664</td>\n",
       "      <td>-0.475790</td>\n",
       "      <td>-1.084986</td>\n",
       "      <td>0.465173</td>\n",
       "      <td>-0.158645</td>\n",
       "      <td>0.098420</td>\n",
       "      <td>-0.481386</td>\n",
       "      <td>-0.070508</td>\n",
       "      <td>-0.015903</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.071417</td>\n",
       "      <td>-0.223865</td>\n",
       "      <td>0.573798</td>\n",
       "      <td>-0.095107</td>\n",
       "      <td>-0.457207</td>\n",
       "      <td>-0.748568</td>\n",
       "      <td>0.635954</td>\n",
       "      <td>-0.240149</td>\n",
       "      <td>0.344973</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.452373</td>\n",
       "      <td>-0.951649</td>\n",
       "      <td>0.070342</td>\n",
       "      <td>-0.817547</td>\n",
       "      <td>-0.840311</td>\n",
       "      <td>0.795529</td>\n",
       "      <td>0.121590</td>\n",
       "      <td>0.601885</td>\n",
       "      <td>-0.231436</td>\n",
       "      <td>-0.664958</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.065126</td>\n",
       "      <td>-0.275068</td>\n",
       "      <td>-0.061993</td>\n",
       "      <td>-0.802680</td>\n",
       "      <td>-0.495667</td>\n",
       "      <td>0.879377</td>\n",
       "      <td>-0.143346</td>\n",
       "      <td>-0.082781</td>\n",
       "      <td>0.353505</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.452337</td>\n",
       "      <td>1.225778</td>\n",
       "      <td>0.626067</td>\n",
       "      <td>-1.426541</td>\n",
       "      <td>-0.249718</td>\n",
       "      <td>-1.193089</td>\n",
       "      <td>-0.654438</td>\n",
       "      <td>1.212763</td>\n",
       "      <td>-0.530135</td>\n",
       "      <td>-0.808822</td>\n",
       "      <td>...</td>\n",
       "      <td>0.110948</td>\n",
       "      <td>-0.758486</td>\n",
       "      <td>1.009918</td>\n",
       "      <td>1.437023</td>\n",
       "      <td>1.576930</td>\n",
       "      <td>-0.812224</td>\n",
       "      <td>-0.513994</td>\n",
       "      <td>0.087050</td>\n",
       "      <td>0.285167</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.452301</td>\n",
       "      <td>-1.045349</td>\n",
       "      <td>0.373695</td>\n",
       "      <td>0.080851</td>\n",
       "      <td>0.870383</td>\n",
       "      <td>0.242749</td>\n",
       "      <td>-0.288918</td>\n",
       "      <td>0.727344</td>\n",
       "      <td>-0.101179</td>\n",
       "      <td>0.657392</td>\n",
       "      <td>...</td>\n",
       "      <td>0.015303</td>\n",
       "      <td>0.215884</td>\n",
       "      <td>0.219736</td>\n",
       "      <td>2.303275</td>\n",
       "      <td>-0.826810</td>\n",
       "      <td>0.574471</td>\n",
       "      <td>-0.041602</td>\n",
       "      <td>0.012498</td>\n",
       "      <td>0.333327</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.452265</td>\n",
       "      <td>-1.006949</td>\n",
       "      <td>0.585796</td>\n",
       "      <td>1.033456</td>\n",
       "      <td>1.288880</td>\n",
       "      <td>-1.089464</td>\n",
       "      <td>-2.431620</td>\n",
       "      <td>0.794830</td>\n",
       "      <td>-0.612916</td>\n",
       "      <td>0.824701</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.201954</td>\n",
       "      <td>-0.053102</td>\n",
       "      <td>0.226051</td>\n",
       "      <td>-1.690609</td>\n",
       "      <td>-1.178848</td>\n",
       "      <td>0.417280</td>\n",
       "      <td>0.019708</td>\n",
       "      <td>-0.069263</td>\n",
       "      <td>0.007530</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1.452265</td>\n",
       "      <td>-1.013382</td>\n",
       "      <td>0.539064</td>\n",
       "      <td>-0.484548</td>\n",
       "      <td>1.452957</td>\n",
       "      <td>1.318202</td>\n",
       "      <td>1.009762</td>\n",
       "      <td>0.684201</td>\n",
       "      <td>0.174900</td>\n",
       "      <td>-0.417668</td>\n",
       "      <td>...</td>\n",
       "      <td>0.412909</td>\n",
       "      <td>0.419496</td>\n",
       "      <td>-0.191922</td>\n",
       "      <td>-1.184522</td>\n",
       "      <td>-0.262500</td>\n",
       "      <td>0.200119</td>\n",
       "      <td>-0.150119</td>\n",
       "      <td>-0.121884</td>\n",
       "      <td>0.276595</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.452265</td>\n",
       "      <td>-0.978689</td>\n",
       "      <td>-0.025438</td>\n",
       "      <td>0.312878</td>\n",
       "      <td>-0.527179</td>\n",
       "      <td>-0.166151</td>\n",
       "      <td>0.023364</td>\n",
       "      <td>-0.034418</td>\n",
       "      <td>-0.064410</td>\n",
       "      <td>-0.090317</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.017233</td>\n",
       "      <td>0.161675</td>\n",
       "      <td>0.413585</td>\n",
       "      <td>1.442761</td>\n",
       "      <td>-1.391687</td>\n",
       "      <td>0.562756</td>\n",
       "      <td>0.026403</td>\n",
       "      <td>0.001606</td>\n",
       "      <td>0.199927</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1.452229</td>\n",
       "      <td>0.844308</td>\n",
       "      <td>-0.591753</td>\n",
       "      <td>-0.085803</td>\n",
       "      <td>-1.356474</td>\n",
       "      <td>-1.318614</td>\n",
       "      <td>-3.558593</td>\n",
       "      <td>0.614877</td>\n",
       "      <td>-1.267491</td>\n",
       "      <td>0.902074</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.271966</td>\n",
       "      <td>-0.658121</td>\n",
       "      <td>0.251007</td>\n",
       "      <td>-1.715494</td>\n",
       "      <td>0.422321</td>\n",
       "      <td>-0.686513</td>\n",
       "      <td>0.109658</td>\n",
       "      <td>-0.216660</td>\n",
       "      <td>0.170368</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1.452229</td>\n",
       "      <td>-0.950241</td>\n",
       "      <td>0.076973</td>\n",
       "      <td>-0.223885</td>\n",
       "      <td>-0.518524</td>\n",
       "      <td>0.382386</td>\n",
       "      <td>0.316344</td>\n",
       "      <td>0.060757</td>\n",
       "      <td>-0.015188</td>\n",
       "      <td>-0.262275</td>\n",
       "      <td>...</td>\n",
       "      <td>0.204789</td>\n",
       "      <td>0.563708</td>\n",
       "      <td>0.035856</td>\n",
       "      <td>-0.575525</td>\n",
       "      <td>-0.712244</td>\n",
       "      <td>-0.512830</td>\n",
       "      <td>0.092233</td>\n",
       "      <td>-0.024531</td>\n",
       "      <td>0.172632</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1.452229</td>\n",
       "      <td>-0.851774</td>\n",
       "      <td>0.276584</td>\n",
       "      <td>-0.014454</td>\n",
       "      <td>-0.943997</td>\n",
       "      <td>0.084479</td>\n",
       "      <td>-0.462501</td>\n",
       "      <td>0.000943</td>\n",
       "      <td>-0.261788</td>\n",
       "      <td>-0.238519</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.076251</td>\n",
       "      <td>-0.006449</td>\n",
       "      <td>0.228713</td>\n",
       "      <td>0.498259</td>\n",
       "      <td>-0.832214</td>\n",
       "      <td>0.644480</td>\n",
       "      <td>-0.040056</td>\n",
       "      <td>-0.073914</td>\n",
       "      <td>-0.244227</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1.452193</td>\n",
       "      <td>1.316418</td>\n",
       "      <td>-0.665085</td>\n",
       "      <td>-0.609705</td>\n",
       "      <td>0.521875</td>\n",
       "      <td>1.240170</td>\n",
       "      <td>0.527401</td>\n",
       "      <td>0.652844</td>\n",
       "      <td>-1.184247</td>\n",
       "      <td>-0.266748</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.178164</td>\n",
       "      <td>-0.391087</td>\n",
       "      <td>0.320458</td>\n",
       "      <td>-0.785823</td>\n",
       "      <td>-0.790447</td>\n",
       "      <td>-2.198904</td>\n",
       "      <td>-0.358870</td>\n",
       "      <td>0.302603</td>\n",
       "      <td>0.156094</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1.452157</td>\n",
       "      <td>1.078804</td>\n",
       "      <td>0.271939</td>\n",
       "      <td>-2.042608</td>\n",
       "      <td>-0.065764</td>\n",
       "      <td>0.171348</td>\n",
       "      <td>-0.446082</td>\n",
       "      <td>0.602331</td>\n",
       "      <td>-0.571258</td>\n",
       "      <td>-1.061541</td>\n",
       "      <td>...</td>\n",
       "      <td>0.120796</td>\n",
       "      <td>-0.015046</td>\n",
       "      <td>0.466788</td>\n",
       "      <td>-0.392280</td>\n",
       "      <td>-0.282131</td>\n",
       "      <td>-0.633253</td>\n",
       "      <td>0.025079</td>\n",
       "      <td>-0.153632</td>\n",
       "      <td>0.263857</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>1.452157</td>\n",
       "      <td>-0.941412</td>\n",
       "      <td>0.285961</td>\n",
       "      <td>0.909578</td>\n",
       "      <td>1.089642</td>\n",
       "      <td>-0.537661</td>\n",
       "      <td>-0.129375</td>\n",
       "      <td>-0.410612</td>\n",
       "      <td>0.075865</td>\n",
       "      <td>-0.593604</td>\n",
       "      <td>...</td>\n",
       "      <td>0.371249</td>\n",
       "      <td>1.402308</td>\n",
       "      <td>0.409415</td>\n",
       "      <td>2.814434</td>\n",
       "      <td>-1.269394</td>\n",
       "      <td>1.930274</td>\n",
       "      <td>-0.047710</td>\n",
       "      <td>-0.050892</td>\n",
       "      <td>-0.243216</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>1.452121</td>\n",
       "      <td>-0.966488</td>\n",
       "      <td>0.738509</td>\n",
       "      <td>-0.506658</td>\n",
       "      <td>0.460218</td>\n",
       "      <td>1.157781</td>\n",
       "      <td>0.009149</td>\n",
       "      <td>1.020093</td>\n",
       "      <td>-0.193854</td>\n",
       "      <td>0.083470</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.517196</td>\n",
       "      <td>-1.237500</td>\n",
       "      <td>0.333368</td>\n",
       "      <td>-0.039436</td>\n",
       "      <td>-0.700985</td>\n",
       "      <td>0.101164</td>\n",
       "      <td>-0.048362</td>\n",
       "      <td>-0.074431</td>\n",
       "      <td>-0.030156</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1.452085</td>\n",
       "      <td>1.398135</td>\n",
       "      <td>2.335082</td>\n",
       "      <td>-1.442586</td>\n",
       "      <td>0.384820</td>\n",
       "      <td>0.069173</td>\n",
       "      <td>0.697448</td>\n",
       "      <td>0.214838</td>\n",
       "      <td>-0.260732</td>\n",
       "      <td>0.474425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.776372</td>\n",
       "      <td>-0.440794</td>\n",
       "      <td>-2.306457</td>\n",
       "      <td>-0.953975</td>\n",
       "      <td>-1.206239</td>\n",
       "      <td>0.686962</td>\n",
       "      <td>-0.186766</td>\n",
       "      <td>-0.698160</td>\n",
       "      <td>-1.664437</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1.452048</td>\n",
       "      <td>0.743035</td>\n",
       "      <td>-0.612449</td>\n",
       "      <td>-1.047389</td>\n",
       "      <td>0.027507</td>\n",
       "      <td>-0.003666</td>\n",
       "      <td>0.474980</td>\n",
       "      <td>-0.587055</td>\n",
       "      <td>-0.029790</td>\n",
       "      <td>0.378340</td>\n",
       "      <td>...</td>\n",
       "      <td>0.219452</td>\n",
       "      <td>0.539188</td>\n",
       "      <td>-0.118283</td>\n",
       "      <td>-0.627893</td>\n",
       "      <td>0.666482</td>\n",
       "      <td>-0.162887</td>\n",
       "      <td>-0.703195</td>\n",
       "      <td>-0.392562</td>\n",
       "      <td>0.311855</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>1.452048</td>\n",
       "      <td>-0.948487</td>\n",
       "      <td>0.058835</td>\n",
       "      <td>0.368274</td>\n",
       "      <td>-0.100133</td>\n",
       "      <td>0.136339</td>\n",
       "      <td>1.057006</td>\n",
       "      <td>-0.548517</td>\n",
       "      <td>0.388016</td>\n",
       "      <td>0.210854</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.050274</td>\n",
       "      <td>0.128409</td>\n",
       "      <td>0.296087</td>\n",
       "      <td>-0.826714</td>\n",
       "      <td>-0.981897</td>\n",
       "      <td>-2.180281</td>\n",
       "      <td>0.329313</td>\n",
       "      <td>-0.035390</td>\n",
       "      <td>-0.122918</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1.452048</td>\n",
       "      <td>-0.908707</td>\n",
       "      <td>0.195007</td>\n",
       "      <td>-0.465370</td>\n",
       "      <td>-1.010256</td>\n",
       "      <td>0.466887</td>\n",
       "      <td>-0.160355</td>\n",
       "      <td>0.255458</td>\n",
       "      <td>-0.127634</td>\n",
       "      <td>-0.843811</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.050922</td>\n",
       "      <td>-0.363765</td>\n",
       "      <td>0.227353</td>\n",
       "      <td>-0.174933</td>\n",
       "      <td>-1.046809</td>\n",
       "      <td>0.528577</td>\n",
       "      <td>-0.145717</td>\n",
       "      <td>-0.116716</td>\n",
       "      <td>0.010199</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>1.452012</td>\n",
       "      <td>-1.000623</td>\n",
       "      <td>-0.206222</td>\n",
       "      <td>0.236789</td>\n",
       "      <td>-0.185509</td>\n",
       "      <td>-0.288243</td>\n",
       "      <td>0.072148</td>\n",
       "      <td>-0.034873</td>\n",
       "      <td>0.028910</td>\n",
       "      <td>0.220763</td>\n",
       "      <td>...</td>\n",
       "      <td>0.397043</td>\n",
       "      <td>1.180666</td>\n",
       "      <td>0.048020</td>\n",
       "      <td>1.434696</td>\n",
       "      <td>-0.660974</td>\n",
       "      <td>-0.302093</td>\n",
       "      <td>0.058534</td>\n",
       "      <td>-0.030147</td>\n",
       "      <td>0.354677</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>1.452012</td>\n",
       "      <td>-0.955614</td>\n",
       "      <td>0.614172</td>\n",
       "      <td>-0.499729</td>\n",
       "      <td>0.596064</td>\n",
       "      <td>0.808809</td>\n",
       "      <td>-0.404591</td>\n",
       "      <td>0.993609</td>\n",
       "      <td>-0.288186</td>\n",
       "      <td>0.569528</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.721183</td>\n",
       "      <td>-1.964235</td>\n",
       "      <td>0.155661</td>\n",
       "      <td>0.384468</td>\n",
       "      <td>-0.375218</td>\n",
       "      <td>0.030722</td>\n",
       "      <td>-0.137716</td>\n",
       "      <td>-0.058033</td>\n",
       "      <td>0.075261</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149970</th>\n",
       "      <td>-1.756428</td>\n",
       "      <td>1.016629</td>\n",
       "      <td>-1.300244</td>\n",
       "      <td>1.647278</td>\n",
       "      <td>0.570551</td>\n",
       "      <td>-0.350534</td>\n",
       "      <td>0.363522</td>\n",
       "      <td>0.061048</td>\n",
       "      <td>-1.010531</td>\n",
       "      <td>0.137970</td>\n",
       "      <td>...</td>\n",
       "      <td>0.442051</td>\n",
       "      <td>1.455620</td>\n",
       "      <td>-0.223045</td>\n",
       "      <td>0.686651</td>\n",
       "      <td>0.264657</td>\n",
       "      <td>-0.387327</td>\n",
       "      <td>-0.172148</td>\n",
       "      <td>0.017392</td>\n",
       "      <td>0.321762</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149971</th>\n",
       "      <td>-1.756428</td>\n",
       "      <td>-1.043101</td>\n",
       "      <td>0.353700</td>\n",
       "      <td>1.752885</td>\n",
       "      <td>-0.701406</td>\n",
       "      <td>-1.040170</td>\n",
       "      <td>-0.820345</td>\n",
       "      <td>-0.539261</td>\n",
       "      <td>-0.166103</td>\n",
       "      <td>0.293354</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.540252</td>\n",
       "      <td>-1.263675</td>\n",
       "      <td>0.168164</td>\n",
       "      <td>2.785013</td>\n",
       "      <td>-0.086130</td>\n",
       "      <td>0.778481</td>\n",
       "      <td>0.018569</td>\n",
       "      <td>0.188774</td>\n",
       "      <td>-0.486967</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149972</th>\n",
       "      <td>-1.756428</td>\n",
       "      <td>0.798367</td>\n",
       "      <td>0.362099</td>\n",
       "      <td>0.977508</td>\n",
       "      <td>0.021173</td>\n",
       "      <td>-0.275638</td>\n",
       "      <td>-0.050383</td>\n",
       "      <td>-0.553856</td>\n",
       "      <td>-0.111420</td>\n",
       "      <td>-0.445095</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.143171</td>\n",
       "      <td>-0.673102</td>\n",
       "      <td>-0.338122</td>\n",
       "      <td>-0.537380</td>\n",
       "      <td>4.372295</td>\n",
       "      <td>0.697698</td>\n",
       "      <td>-0.428510</td>\n",
       "      <td>0.611492</td>\n",
       "      <td>-0.424574</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149973</th>\n",
       "      <td>-1.756465</td>\n",
       "      <td>0.913170</td>\n",
       "      <td>-0.575903</td>\n",
       "      <td>0.139076</td>\n",
       "      <td>0.844986</td>\n",
       "      <td>-0.619400</td>\n",
       "      <td>0.592743</td>\n",
       "      <td>-0.788683</td>\n",
       "      <td>0.047656</td>\n",
       "      <td>0.262468</td>\n",
       "      <td>...</td>\n",
       "      <td>0.105658</td>\n",
       "      <td>0.376347</td>\n",
       "      <td>-0.060735</td>\n",
       "      <td>0.530556</td>\n",
       "      <td>1.840479</td>\n",
       "      <td>0.199816</td>\n",
       "      <td>1.408987</td>\n",
       "      <td>0.240229</td>\n",
       "      <td>0.321924</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149974</th>\n",
       "      <td>-1.756465</td>\n",
       "      <td>0.753591</td>\n",
       "      <td>-0.179997</td>\n",
       "      <td>-1.012101</td>\n",
       "      <td>0.599390</td>\n",
       "      <td>-0.437494</td>\n",
       "      <td>-0.826563</td>\n",
       "      <td>-0.194584</td>\n",
       "      <td>-0.283565</td>\n",
       "      <td>-0.656740</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.322517</td>\n",
       "      <td>-1.683961</td>\n",
       "      <td>0.222540</td>\n",
       "      <td>1.640831</td>\n",
       "      <td>1.254492</td>\n",
       "      <td>-1.287262</td>\n",
       "      <td>-0.637784</td>\n",
       "      <td>0.051539</td>\n",
       "      <td>0.311774</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149975</th>\n",
       "      <td>-1.756501</td>\n",
       "      <td>0.496720</td>\n",
       "      <td>-0.225179</td>\n",
       "      <td>-0.534901</td>\n",
       "      <td>0.678979</td>\n",
       "      <td>-0.185976</td>\n",
       "      <td>0.008957</td>\n",
       "      <td>-0.360932</td>\n",
       "      <td>-0.150653</td>\n",
       "      <td>-0.042370</td>\n",
       "      <td>...</td>\n",
       "      <td>0.051480</td>\n",
       "      <td>0.280999</td>\n",
       "      <td>-0.567845</td>\n",
       "      <td>-1.220672</td>\n",
       "      <td>2.953577</td>\n",
       "      <td>-0.010524</td>\n",
       "      <td>-0.445146</td>\n",
       "      <td>-0.783423</td>\n",
       "      <td>0.209066</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149976</th>\n",
       "      <td>-1.756537</td>\n",
       "      <td>-0.466915</td>\n",
       "      <td>-0.586969</td>\n",
       "      <td>0.575110</td>\n",
       "      <td>0.585976</td>\n",
       "      <td>-0.949058</td>\n",
       "      <td>0.174812</td>\n",
       "      <td>-0.742782</td>\n",
       "      <td>-0.061101</td>\n",
       "      <td>0.200091</td>\n",
       "      <td>...</td>\n",
       "      <td>0.426840</td>\n",
       "      <td>1.139958</td>\n",
       "      <td>-0.051097</td>\n",
       "      <td>0.032106</td>\n",
       "      <td>0.825265</td>\n",
       "      <td>-0.263748</td>\n",
       "      <td>-0.541541</td>\n",
       "      <td>-0.219618</td>\n",
       "      <td>0.334055</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149977</th>\n",
       "      <td>-1.756537</td>\n",
       "      <td>-1.159550</td>\n",
       "      <td>0.170874</td>\n",
       "      <td>0.917977</td>\n",
       "      <td>-0.387761</td>\n",
       "      <td>0.133238</td>\n",
       "      <td>0.688645</td>\n",
       "      <td>-0.023712</td>\n",
       "      <td>0.180386</td>\n",
       "      <td>-1.117878</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.137221</td>\n",
       "      <td>-0.821826</td>\n",
       "      <td>-0.040305</td>\n",
       "      <td>0.101618</td>\n",
       "      <td>-0.405998</td>\n",
       "      <td>0.281043</td>\n",
       "      <td>0.010296</td>\n",
       "      <td>0.192963</td>\n",
       "      <td>0.302757</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149978</th>\n",
       "      <td>-1.756573</td>\n",
       "      <td>-1.059527</td>\n",
       "      <td>0.689100</td>\n",
       "      <td>1.290097</td>\n",
       "      <td>-0.255152</td>\n",
       "      <td>-1.236183</td>\n",
       "      <td>-3.189057</td>\n",
       "      <td>0.813775</td>\n",
       "      <td>-0.892983</td>\n",
       "      <td>-0.837991</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.132336</td>\n",
       "      <td>0.185648</td>\n",
       "      <td>-0.293893</td>\n",
       "      <td>-1.078237</td>\n",
       "      <td>0.835051</td>\n",
       "      <td>1.706793</td>\n",
       "      <td>-0.131372</td>\n",
       "      <td>-0.018202</td>\n",
       "      <td>-0.471076</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149979</th>\n",
       "      <td>-1.756573</td>\n",
       "      <td>1.182346</td>\n",
       "      <td>0.966613</td>\n",
       "      <td>0.162129</td>\n",
       "      <td>1.441268</td>\n",
       "      <td>-1.106665</td>\n",
       "      <td>0.459592</td>\n",
       "      <td>-0.037243</td>\n",
       "      <td>0.008725</td>\n",
       "      <td>0.883619</td>\n",
       "      <td>...</td>\n",
       "      <td>0.053925</td>\n",
       "      <td>0.095853</td>\n",
       "      <td>0.832632</td>\n",
       "      <td>2.361111</td>\n",
       "      <td>-2.101367</td>\n",
       "      <td>0.152280</td>\n",
       "      <td>-0.593383</td>\n",
       "      <td>0.037510</td>\n",
       "      <td>-0.573582</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149980</th>\n",
       "      <td>-1.756573</td>\n",
       "      <td>-0.391201</td>\n",
       "      <td>-0.612489</td>\n",
       "      <td>0.372026</td>\n",
       "      <td>0.482981</td>\n",
       "      <td>-0.761526</td>\n",
       "      <td>0.391464</td>\n",
       "      <td>-0.767989</td>\n",
       "      <td>-0.026102</td>\n",
       "      <td>0.304215</td>\n",
       "      <td>...</td>\n",
       "      <td>0.352909</td>\n",
       "      <td>0.881874</td>\n",
       "      <td>-0.184893</td>\n",
       "      <td>-0.999227</td>\n",
       "      <td>0.999803</td>\n",
       "      <td>-0.186210</td>\n",
       "      <td>-0.559111</td>\n",
       "      <td>-0.242803</td>\n",
       "      <td>0.334055</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149981</th>\n",
       "      <td>-1.756609</td>\n",
       "      <td>-1.147477</td>\n",
       "      <td>0.279278</td>\n",
       "      <td>0.281969</td>\n",
       "      <td>-0.345473</td>\n",
       "      <td>0.565136</td>\n",
       "      <td>0.462859</td>\n",
       "      <td>0.455514</td>\n",
       "      <td>0.103554</td>\n",
       "      <td>-1.049987</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.338354</td>\n",
       "      <td>-1.350828</td>\n",
       "      <td>-0.172678</td>\n",
       "      <td>-0.227034</td>\n",
       "      <td>0.482992</td>\n",
       "      <td>-1.278385</td>\n",
       "      <td>0.043991</td>\n",
       "      <td>0.140380</td>\n",
       "      <td>0.237007</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149982</th>\n",
       "      <td>-1.756645</td>\n",
       "      <td>1.165315</td>\n",
       "      <td>-1.533038</td>\n",
       "      <td>1.598313</td>\n",
       "      <td>0.394605</td>\n",
       "      <td>0.319690</td>\n",
       "      <td>0.755502</td>\n",
       "      <td>0.317364</td>\n",
       "      <td>-1.255737</td>\n",
       "      <td>0.179836</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.621675</td>\n",
       "      <td>-1.788662</td>\n",
       "      <td>-0.210192</td>\n",
       "      <td>-1.177607</td>\n",
       "      <td>1.078993</td>\n",
       "      <td>0.471305</td>\n",
       "      <td>-0.847496</td>\n",
       "      <td>-0.804946</td>\n",
       "      <td>0.323946</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149983</th>\n",
       "      <td>-1.756681</td>\n",
       "      <td>-1.119633</td>\n",
       "      <td>0.048303</td>\n",
       "      <td>1.680693</td>\n",
       "      <td>-0.843280</td>\n",
       "      <td>-0.628014</td>\n",
       "      <td>0.686008</td>\n",
       "      <td>-0.810070</td>\n",
       "      <td>0.374889</td>\n",
       "      <td>0.138759</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.281208</td>\n",
       "      <td>-0.521394</td>\n",
       "      <td>0.165646</td>\n",
       "      <td>-1.000763</td>\n",
       "      <td>-0.810031</td>\n",
       "      <td>1.082404</td>\n",
       "      <td>0.152179</td>\n",
       "      <td>0.113169</td>\n",
       "      <td>-0.224009</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149984</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.449607</td>\n",
       "      <td>-0.193735</td>\n",
       "      <td>-0.449793</td>\n",
       "      <td>-0.362727</td>\n",
       "      <td>0.273393</td>\n",
       "      <td>-1.050009</td>\n",
       "      <td>-0.940083</td>\n",
       "      <td>0.474071</td>\n",
       "      <td>-0.620439</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.187227</td>\n",
       "      <td>-1.684362</td>\n",
       "      <td>0.254510</td>\n",
       "      <td>-0.140337</td>\n",
       "      <td>1.753141</td>\n",
       "      <td>-0.855840</td>\n",
       "      <td>0.447327</td>\n",
       "      <td>0.930464</td>\n",
       "      <td>-0.450453</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149985</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.383014</td>\n",
       "      <td>-0.643444</td>\n",
       "      <td>0.903562</td>\n",
       "      <td>0.007353</td>\n",
       "      <td>-0.176866</td>\n",
       "      <td>1.013712</td>\n",
       "      <td>-0.805823</td>\n",
       "      <td>0.000068</td>\n",
       "      <td>0.325499</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.538168</td>\n",
       "      <td>-1.524937</td>\n",
       "      <td>-0.125802</td>\n",
       "      <td>-0.001612</td>\n",
       "      <td>0.454743</td>\n",
       "      <td>0.331630</td>\n",
       "      <td>0.218033</td>\n",
       "      <td>0.127822</td>\n",
       "      <td>0.013555</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149986</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.964503</td>\n",
       "      <td>1.117810</td>\n",
       "      <td>0.239642</td>\n",
       "      <td>-0.738516</td>\n",
       "      <td>0.994184</td>\n",
       "      <td>-0.236225</td>\n",
       "      <td>0.394180</td>\n",
       "      <td>-0.140536</td>\n",
       "      <td>-1.340730</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.133608</td>\n",
       "      <td>0.306317</td>\n",
       "      <td>-0.022101</td>\n",
       "      <td>-0.084473</td>\n",
       "      <td>1.341826</td>\n",
       "      <td>-0.126860</td>\n",
       "      <td>0.129507</td>\n",
       "      <td>-0.052676</td>\n",
       "      <td>-1.173457</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149987</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.513373</td>\n",
       "      <td>-0.677014</td>\n",
       "      <td>0.863835</td>\n",
       "      <td>0.024536</td>\n",
       "      <td>-0.201272</td>\n",
       "      <td>1.122639</td>\n",
       "      <td>-0.526677</td>\n",
       "      <td>-0.095007</td>\n",
       "      <td>0.076104</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.461505</td>\n",
       "      <td>-1.339734</td>\n",
       "      <td>0.083941</td>\n",
       "      <td>0.060320</td>\n",
       "      <td>0.617299</td>\n",
       "      <td>0.315958</td>\n",
       "      <td>0.189216</td>\n",
       "      <td>0.154492</td>\n",
       "      <td>0.272227</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149988</th>\n",
       "      <td>-1.756753</td>\n",
       "      <td>1.018889</td>\n",
       "      <td>-0.875585</td>\n",
       "      <td>0.468502</td>\n",
       "      <td>1.310742</td>\n",
       "      <td>-0.921393</td>\n",
       "      <td>0.088205</td>\n",
       "      <td>-0.842800</td>\n",
       "      <td>-0.111500</td>\n",
       "      <td>0.088516</td>\n",
       "      <td>...</td>\n",
       "      <td>0.046710</td>\n",
       "      <td>0.536748</td>\n",
       "      <td>0.356054</td>\n",
       "      <td>2.294246</td>\n",
       "      <td>0.154244</td>\n",
       "      <td>1.656127</td>\n",
       "      <td>1.730489</td>\n",
       "      <td>-0.338444</td>\n",
       "      <td>0.310642</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149989</th>\n",
       "      <td>-1.756753</td>\n",
       "      <td>-1.128032</td>\n",
       "      <td>0.241159</td>\n",
       "      <td>0.848114</td>\n",
       "      <td>-0.346470</td>\n",
       "      <td>0.184799</td>\n",
       "      <td>0.745318</td>\n",
       "      <td>-0.136615</td>\n",
       "      <td>0.250803</td>\n",
       "      <td>-0.799592</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.111768</td>\n",
       "      <td>-0.377428</td>\n",
       "      <td>-0.141664</td>\n",
       "      <td>-0.008585</td>\n",
       "      <td>0.118576</td>\n",
       "      <td>0.721861</td>\n",
       "      <td>0.022403</td>\n",
       "      <td>0.109630</td>\n",
       "      <td>-0.020411</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149990</th>\n",
       "      <td>-1.756789</td>\n",
       "      <td>0.720683</td>\n",
       "      <td>-0.697570</td>\n",
       "      <td>-0.215227</td>\n",
       "      <td>-0.409272</td>\n",
       "      <td>-0.783372</td>\n",
       "      <td>0.368362</td>\n",
       "      <td>-1.490595</td>\n",
       "      <td>0.258694</td>\n",
       "      <td>0.989876</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.146275</td>\n",
       "      <td>-0.474964</td>\n",
       "      <td>0.566948</td>\n",
       "      <td>-1.845373</td>\n",
       "      <td>-2.557826</td>\n",
       "      <td>0.708178</td>\n",
       "      <td>0.210015</td>\n",
       "      <td>0.187881</td>\n",
       "      <td>0.067942</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149991</th>\n",
       "      <td>-1.756789</td>\n",
       "      <td>0.843747</td>\n",
       "      <td>-0.297288</td>\n",
       "      <td>-0.037355</td>\n",
       "      <td>-0.909128</td>\n",
       "      <td>0.373785</td>\n",
       "      <td>-0.638166</td>\n",
       "      <td>-0.733955</td>\n",
       "      <td>-0.311541</td>\n",
       "      <td>-0.060952</td>\n",
       "      <td>...</td>\n",
       "      <td>0.512972</td>\n",
       "      <td>1.338080</td>\n",
       "      <td>-0.840501</td>\n",
       "      <td>-0.910830</td>\n",
       "      <td>1.219112</td>\n",
       "      <td>2.135645</td>\n",
       "      <td>-0.031636</td>\n",
       "      <td>0.091237</td>\n",
       "      <td>-0.532013</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149992</th>\n",
       "      <td>-1.756789</td>\n",
       "      <td>-1.158497</td>\n",
       "      <td>0.080272</td>\n",
       "      <td>1.628651</td>\n",
       "      <td>-0.077196</td>\n",
       "      <td>-0.490708</td>\n",
       "      <td>0.693930</td>\n",
       "      <td>-0.519585</td>\n",
       "      <td>0.275900</td>\n",
       "      <td>0.047585</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.239883</td>\n",
       "      <td>-0.763969</td>\n",
       "      <td>0.043544</td>\n",
       "      <td>-1.400203</td>\n",
       "      <td>-0.556032</td>\n",
       "      <td>-1.583964</td>\n",
       "      <td>0.352115</td>\n",
       "      <td>0.270728</td>\n",
       "      <td>0.174856</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149993</th>\n",
       "      <td>-1.756825</td>\n",
       "      <td>-1.163542</td>\n",
       "      <td>0.253798</td>\n",
       "      <td>2.003056</td>\n",
       "      <td>0.513224</td>\n",
       "      <td>-1.844906</td>\n",
       "      <td>-2.527033</td>\n",
       "      <td>0.338701</td>\n",
       "      <td>-0.642826</td>\n",
       "      <td>-0.424007</td>\n",
       "      <td>...</td>\n",
       "      <td>0.321263</td>\n",
       "      <td>0.925534</td>\n",
       "      <td>-0.520084</td>\n",
       "      <td>-1.171770</td>\n",
       "      <td>0.392388</td>\n",
       "      <td>-0.610395</td>\n",
       "      <td>0.143045</td>\n",
       "      <td>0.226016</td>\n",
       "      <td>0.310440</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149994</th>\n",
       "      <td>-1.756825</td>\n",
       "      <td>-1.165533</td>\n",
       "      <td>0.935357</td>\n",
       "      <td>1.386949</td>\n",
       "      <td>1.325895</td>\n",
       "      <td>-0.858063</td>\n",
       "      <td>-2.924312</td>\n",
       "      <td>1.457903</td>\n",
       "      <td>-0.835015</td>\n",
       "      <td>-0.189523</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.633114</td>\n",
       "      <td>-1.798527</td>\n",
       "      <td>-0.132122</td>\n",
       "      <td>-1.254339</td>\n",
       "      <td>0.239633</td>\n",
       "      <td>-0.079566</td>\n",
       "      <td>-0.086058</td>\n",
       "      <td>0.147121</td>\n",
       "      <td>0.116547</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149995</th>\n",
       "      <td>-1.756898</td>\n",
       "      <td>0.867638</td>\n",
       "      <td>-0.905587</td>\n",
       "      <td>1.055569</td>\n",
       "      <td>0.517767</td>\n",
       "      <td>-0.276563</td>\n",
       "      <td>0.266143</td>\n",
       "      <td>-0.074745</td>\n",
       "      <td>-0.774578</td>\n",
       "      <td>0.146479</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.592917</td>\n",
       "      <td>-1.703160</td>\n",
       "      <td>0.030564</td>\n",
       "      <td>-0.349722</td>\n",
       "      <td>1.462083</td>\n",
       "      <td>0.469627</td>\n",
       "      <td>-0.878303</td>\n",
       "      <td>-0.798875</td>\n",
       "      <td>0.354273</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149996</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>0.415722</td>\n",
       "      <td>-1.200718</td>\n",
       "      <td>-0.259352</td>\n",
       "      <td>-2.851478</td>\n",
       "      <td>-0.421152</td>\n",
       "      <td>-0.370122</td>\n",
       "      <td>-2.084808</td>\n",
       "      <td>1.104757</td>\n",
       "      <td>0.490121</td>\n",
       "      <td>...</td>\n",
       "      <td>0.084737</td>\n",
       "      <td>-1.981259</td>\n",
       "      <td>0.301342</td>\n",
       "      <td>-0.115487</td>\n",
       "      <td>2.360976</td>\n",
       "      <td>-0.549872</td>\n",
       "      <td>-0.711672</td>\n",
       "      <td>1.911911</td>\n",
       "      <td>-0.254902</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149997</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>1.167842</td>\n",
       "      <td>-0.552907</td>\n",
       "      <td>0.423081</td>\n",
       "      <td>0.155091</td>\n",
       "      <td>-0.270602</td>\n",
       "      <td>0.121991</td>\n",
       "      <td>-1.100427</td>\n",
       "      <td>0.400494</td>\n",
       "      <td>-0.304126</td>\n",
       "      <td>...</td>\n",
       "      <td>0.448856</td>\n",
       "      <td>0.183284</td>\n",
       "      <td>0.069664</td>\n",
       "      <td>0.705520</td>\n",
       "      <td>1.235442</td>\n",
       "      <td>-0.582444</td>\n",
       "      <td>1.837158</td>\n",
       "      <td>0.510018</td>\n",
       "      <td>-0.079287</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149998</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>0.969063</td>\n",
       "      <td>-0.579037</td>\n",
       "      <td>-1.038397</td>\n",
       "      <td>0.288642</td>\n",
       "      <td>0.177344</td>\n",
       "      <td>-0.353638</td>\n",
       "      <td>0.066474</td>\n",
       "      <td>-0.686863</td>\n",
       "      <td>-0.341261</td>\n",
       "      <td>...</td>\n",
       "      <td>0.470687</td>\n",
       "      <td>1.151563</td>\n",
       "      <td>0.396852</td>\n",
       "      <td>0.848703</td>\n",
       "      <td>-0.831973</td>\n",
       "      <td>0.858601</td>\n",
       "      <td>-0.507758</td>\n",
       "      <td>-0.172860</td>\n",
       "      <td>0.354233</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149999</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>-1.216907</td>\n",
       "      <td>0.306214</td>\n",
       "      <td>1.981126</td>\n",
       "      <td>0.715126</td>\n",
       "      <td>-0.286982</td>\n",
       "      <td>1.128890</td>\n",
       "      <td>-0.298423</td>\n",
       "      <td>0.434975</td>\n",
       "      <td>0.752355</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.625348</td>\n",
       "      <td>-1.911075</td>\n",
       "      <td>0.259232</td>\n",
       "      <td>-1.212488</td>\n",
       "      <td>-1.381696</td>\n",
       "      <td>-0.572706</td>\n",
       "      <td>0.263233</td>\n",
       "      <td>0.301439</td>\n",
       "      <td>0.351038</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150000 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Ocorrencia       PP1       PP2       PP3       PP4       PP5  \\\n",
       "0         1.452554 -0.985401  0.607243 -0.578142  0.206595  0.604854   \n",
       "1         1.452518 -0.711537  1.151666 -0.060850  0.830176  1.091942   \n",
       "2         1.452481  0.615415  2.128527 -0.377129 -0.279669  1.826966   \n",
       "3         1.452481 -0.935186 -0.064476 -0.024557 -0.638310 -0.120940   \n",
       "4         1.452445  1.555368  1.615105  0.756921  0.294379  1.039644   \n",
       "5         1.452445 -0.981881 -0.034877 -0.322386 -0.838697  0.332925   \n",
       "6         1.452445 -0.971023 -0.174094  0.022727 -0.596579 -0.163207   \n",
       "7         1.452445  1.422906  2.943124 -0.502896  0.271327 -2.055703   \n",
       "8         1.452445 -0.994214  0.265920 -0.129913  1.049748  0.553487   \n",
       "9         1.452409  0.893312  0.262019 -0.938725  1.166103  0.781218   \n",
       "10        1.452373  0.909664 -0.475790 -1.084986  0.465173 -0.158645   \n",
       "11        1.452373 -0.951649  0.070342 -0.817547 -0.840311  0.795529   \n",
       "12        1.452337  1.225778  0.626067 -1.426541 -0.249718 -1.193089   \n",
       "13        1.452301 -1.045349  0.373695  0.080851  0.870383  0.242749   \n",
       "14        1.452265 -1.006949  0.585796  1.033456  1.288880 -1.089464   \n",
       "15        1.452265 -1.013382  0.539064 -0.484548  1.452957  1.318202   \n",
       "16        1.452265 -0.978689 -0.025438  0.312878 -0.527179 -0.166151   \n",
       "17        1.452229  0.844308 -0.591753 -0.085803 -1.356474 -1.318614   \n",
       "18        1.452229 -0.950241  0.076973 -0.223885 -0.518524  0.382386   \n",
       "19        1.452229 -0.851774  0.276584 -0.014454 -0.943997  0.084479   \n",
       "20        1.452193  1.316418 -0.665085 -0.609705  0.521875  1.240170   \n",
       "21        1.452157  1.078804  0.271939 -2.042608 -0.065764  0.171348   \n",
       "22        1.452157 -0.941412  0.285961  0.909578  1.089642 -0.537661   \n",
       "23        1.452121 -0.966488  0.738509 -0.506658  0.460218  1.157781   \n",
       "24        1.452085  1.398135  2.335082 -1.442586  0.384820  0.069173   \n",
       "25        1.452048  0.743035 -0.612449 -1.047389  0.027507 -0.003666   \n",
       "26        1.452048 -0.948487  0.058835  0.368274 -0.100133  0.136339   \n",
       "27        1.452048 -0.908707  0.195007 -0.465370 -1.010256  0.466887   \n",
       "28        1.452012 -1.000623 -0.206222  0.236789 -0.185509 -0.288243   \n",
       "29        1.452012 -0.955614  0.614172 -0.499729  0.596064  0.808809   \n",
       "...            ...       ...       ...       ...       ...       ...   \n",
       "149970   -1.756428  1.016629 -1.300244  1.647278  0.570551 -0.350534   \n",
       "149971   -1.756428 -1.043101  0.353700  1.752885 -0.701406 -1.040170   \n",
       "149972   -1.756428  0.798367  0.362099  0.977508  0.021173 -0.275638   \n",
       "149973   -1.756465  0.913170 -0.575903  0.139076  0.844986 -0.619400   \n",
       "149974   -1.756465  0.753591 -0.179997 -1.012101  0.599390 -0.437494   \n",
       "149975   -1.756501  0.496720 -0.225179 -0.534901  0.678979 -0.185976   \n",
       "149976   -1.756537 -0.466915 -0.586969  0.575110  0.585976 -0.949058   \n",
       "149977   -1.756537 -1.159550  0.170874  0.917977 -0.387761  0.133238   \n",
       "149978   -1.756573 -1.059527  0.689100  1.290097 -0.255152 -1.236183   \n",
       "149979   -1.756573  1.182346  0.966613  0.162129  1.441268 -1.106665   \n",
       "149980   -1.756573 -0.391201 -0.612489  0.372026  0.482981 -0.761526   \n",
       "149981   -1.756609 -1.147477  0.279278  0.281969 -0.345473  0.565136   \n",
       "149982   -1.756645  1.165315 -1.533038  1.598313  0.394605  0.319690   \n",
       "149983   -1.756681 -1.119633  0.048303  1.680693 -0.843280 -0.628014   \n",
       "149984   -1.756717 -0.449607 -0.193735 -0.449793 -0.362727  0.273393   \n",
       "149985   -1.756717 -0.383014 -0.643444  0.903562  0.007353 -0.176866   \n",
       "149986   -1.756717 -0.964503  1.117810  0.239642 -0.738516  0.994184   \n",
       "149987   -1.756717 -0.513373 -0.677014  0.863835  0.024536 -0.201272   \n",
       "149988   -1.756753  1.018889 -0.875585  0.468502  1.310742 -0.921393   \n",
       "149989   -1.756753 -1.128032  0.241159  0.848114 -0.346470  0.184799   \n",
       "149990   -1.756789  0.720683 -0.697570 -0.215227 -0.409272 -0.783372   \n",
       "149991   -1.756789  0.843747 -0.297288 -0.037355 -0.909128  0.373785   \n",
       "149992   -1.756789 -1.158497  0.080272  1.628651 -0.077196 -0.490708   \n",
       "149993   -1.756825 -1.163542  0.253798  2.003056  0.513224 -1.844906   \n",
       "149994   -1.756825 -1.165533  0.935357  1.386949  1.325895 -0.858063   \n",
       "149995   -1.756898  0.867638 -0.905587  1.055569  0.517767 -0.276563   \n",
       "149996   -1.756934  0.415722 -1.200718 -0.259352 -2.851478 -0.421152   \n",
       "149997   -1.756934  1.167842 -0.552907  0.423081  0.155091 -0.270602   \n",
       "149998   -1.756934  0.969063 -0.579037 -1.038397  0.288642  0.177344   \n",
       "149999   -1.756934 -1.216907  0.306214  1.981126  0.715126 -0.286982   \n",
       "\n",
       "             PP6       PP7       PP8       PP9   ...        PP21      PP22  \\\n",
       "0      -1.260358  1.192228 -0.393617 -0.033998   ...    0.510952  0.365689   \n",
       "1       0.219741  0.180462 -0.001535 -1.525840   ...   -0.537851 -0.652846   \n",
       "2      -0.392920  0.266475 -0.007794 -0.236547   ...   -0.865822 -0.587094   \n",
       "3      -0.209379  0.002692 -0.167246  0.185366   ...   -0.212839 -0.641821   \n",
       "4      -1.149025 -0.158615 -1.100691 -0.333398   ...    0.285680  1.097367   \n",
       "5       0.196228  0.192049 -0.057406 -0.753223   ...    0.516861  1.400395   \n",
       "6       0.004658 -0.057781 -0.047207  0.239018   ...   -0.129952 -0.481487   \n",
       "7       0.641306  1.064781 -0.404160  1.005183   ...   -0.333678  0.649857   \n",
       "8       0.475634  0.293952 -0.014615 -1.404025   ...    0.426987  0.975534   \n",
       "9       0.405424 -0.717286  0.049532  1.476413   ...   -0.036519 -0.132637   \n",
       "10      0.098420 -0.481386 -0.070508 -0.015903   ...   -0.071417 -0.223865   \n",
       "11      0.121590  0.601885 -0.231436 -0.664958   ...   -0.065126 -0.275068   \n",
       "12     -0.654438  1.212763 -0.530135 -0.808822   ...    0.110948 -0.758486   \n",
       "13     -0.288918  0.727344 -0.101179  0.657392   ...    0.015303  0.215884   \n",
       "14     -2.431620  0.794830 -0.612916  0.824701   ...   -0.201954 -0.053102   \n",
       "15      1.009762  0.684201  0.174900 -0.417668   ...    0.412909  0.419496   \n",
       "16      0.023364 -0.034418 -0.064410 -0.090317   ...   -0.017233  0.161675   \n",
       "17     -3.558593  0.614877 -1.267491  0.902074   ...   -0.271966 -0.658121   \n",
       "18      0.316344  0.060757 -0.015188 -0.262275   ...    0.204789  0.563708   \n",
       "19     -0.462501  0.000943 -0.261788 -0.238519   ...   -0.076251 -0.006449   \n",
       "20      0.527401  0.652844 -1.184247 -0.266748   ...   -0.178164 -0.391087   \n",
       "21     -0.446082  0.602331 -0.571258 -1.061541   ...    0.120796 -0.015046   \n",
       "22     -0.129375 -0.410612  0.075865 -0.593604   ...    0.371249  1.402308   \n",
       "23      0.009149  1.020093 -0.193854  0.083470   ...   -0.517196 -1.237500   \n",
       "24      0.697448  0.214838 -0.260732  0.474425   ...   -0.776372 -0.440794   \n",
       "25      0.474980 -0.587055 -0.029790  0.378340   ...    0.219452  0.539188   \n",
       "26      1.057006 -0.548517  0.388016  0.210854   ...   -0.050274  0.128409   \n",
       "27     -0.160355  0.255458 -0.127634 -0.843811   ...   -0.050922 -0.363765   \n",
       "28      0.072148 -0.034873  0.028910  0.220763   ...    0.397043  1.180666   \n",
       "29     -0.404591  0.993609 -0.288186  0.569528   ...   -0.721183 -1.964235   \n",
       "...          ...       ...       ...       ...   ...         ...       ...   \n",
       "149970  0.363522  0.061048 -1.010531  0.137970   ...    0.442051  1.455620   \n",
       "149971 -0.820345 -0.539261 -0.166103  0.293354   ...   -0.540252 -1.263675   \n",
       "149972 -0.050383 -0.553856 -0.111420 -0.445095   ...   -0.143171 -0.673102   \n",
       "149973  0.592743 -0.788683  0.047656  0.262468   ...    0.105658  0.376347   \n",
       "149974 -0.826563 -0.194584 -0.283565 -0.656740   ...   -0.322517 -1.683961   \n",
       "149975  0.008957 -0.360932 -0.150653 -0.042370   ...    0.051480  0.280999   \n",
       "149976  0.174812 -0.742782 -0.061101  0.200091   ...    0.426840  1.139958   \n",
       "149977  0.688645 -0.023712  0.180386 -1.117878   ...   -0.137221 -0.821826   \n",
       "149978 -3.189057  0.813775 -0.892983 -0.837991   ...   -0.132336  0.185648   \n",
       "149979  0.459592 -0.037243  0.008725  0.883619   ...    0.053925  0.095853   \n",
       "149980  0.391464 -0.767989 -0.026102  0.304215   ...    0.352909  0.881874   \n",
       "149981  0.462859  0.455514  0.103554 -1.049987   ...   -0.338354 -1.350828   \n",
       "149982  0.755502  0.317364 -1.255737  0.179836   ...   -0.621675 -1.788662   \n",
       "149983  0.686008 -0.810070  0.374889  0.138759   ...   -0.281208 -0.521394   \n",
       "149984 -1.050009 -0.940083  0.474071 -0.620439   ...   -0.187227 -1.684362   \n",
       "149985  1.013712 -0.805823  0.000068  0.325499   ...   -0.538168 -1.524937   \n",
       "149986 -0.236225  0.394180 -0.140536 -1.340730   ...   -0.133608  0.306317   \n",
       "149987  1.122639 -0.526677 -0.095007  0.076104   ...   -0.461505 -1.339734   \n",
       "149988  0.088205 -0.842800 -0.111500  0.088516   ...    0.046710  0.536748   \n",
       "149989  0.745318 -0.136615  0.250803 -0.799592   ...   -0.111768 -0.377428   \n",
       "149990  0.368362 -1.490595  0.258694  0.989876   ...   -0.146275 -0.474964   \n",
       "149991 -0.638166 -0.733955 -0.311541 -0.060952   ...    0.512972  1.338080   \n",
       "149992  0.693930 -0.519585  0.275900  0.047585   ...   -0.239883 -0.763969   \n",
       "149993 -2.527033  0.338701 -0.642826 -0.424007   ...    0.321263  0.925534   \n",
       "149994 -2.924312  1.457903 -0.835015 -0.189523   ...   -0.633114 -1.798527   \n",
       "149995  0.266143 -0.074745 -0.774578  0.146479   ...   -0.592917 -1.703160   \n",
       "149996 -0.370122 -2.084808  1.104757  0.490121   ...    0.084737 -1.981259   \n",
       "149997  0.121991 -1.100427  0.400494 -0.304126   ...    0.448856  0.183284   \n",
       "149998 -0.353638  0.066474 -0.686863 -0.341261   ...    0.470687  1.151563   \n",
       "149999  1.128890 -0.298423  0.434975  0.752355   ...   -0.625348 -1.911075   \n",
       "\n",
       "            PP23      PP24      PP25      PP26      PP27      PP28    Sacado  \\\n",
       "0       0.195001  2.114023 -0.525371 -1.389421 -0.173706 -0.027852  0.243518   \n",
       "1       0.811936 -0.178356 -0.623881 -0.150271  0.089540 -0.257123 -1.287488   \n",
       "2       1.108066 -0.857295 -0.068399  0.408465  0.132474 -0.538396 -2.876641   \n",
       "3       0.101408  0.470917 -0.894097  0.598090 -0.088287 -0.020401  0.231791   \n",
       "4       2.090826  2.484221  0.583625 -1.778647 -0.326980  3.330732 -1.753154   \n",
       "5      -0.121911  0.317786 -0.708302  1.250213 -0.097305 -0.074044  0.349097   \n",
       "6       0.185340  0.481784 -1.189520  0.614738 -0.069092 -0.007839  0.352211   \n",
       "7      -2.266220  2.286027  0.180772 -0.699595  0.306683 -0.735704 -1.198487   \n",
       "8      -0.086998 -0.036538 -0.642520  0.584208 -0.051309 -0.015922  0.353666   \n",
       "9      -0.447360 -0.906489  0.154585 -2.006428  0.161230 -0.315946 -0.464282   \n",
       "10      0.573798 -0.095107 -0.457207 -0.748568  0.635954 -0.240149  0.344973   \n",
       "11     -0.061993 -0.802680 -0.495667  0.879377 -0.143346 -0.082781  0.353505   \n",
       "12      1.009918  1.437023  1.576930 -0.812224 -0.513994  0.087050  0.285167   \n",
       "13      0.219736  2.303275 -0.826810  0.574471 -0.041602  0.012498  0.333327   \n",
       "14      0.226051 -1.690609 -1.178848  0.417280  0.019708 -0.069263  0.007530   \n",
       "15     -0.191922 -1.184522 -0.262500  0.200119 -0.150119 -0.121884  0.276595   \n",
       "16      0.413585  1.442761 -1.391687  0.562756  0.026403  0.001606  0.199927   \n",
       "17      0.251007 -1.715494  0.422321 -0.686513  0.109658 -0.216660  0.170368   \n",
       "18      0.035856 -0.575525 -0.712244 -0.512830  0.092233 -0.024531  0.172632   \n",
       "19      0.228713  0.498259 -0.832214  0.644480 -0.040056 -0.073914 -0.244227   \n",
       "20      0.320458 -0.785823 -0.790447 -2.198904 -0.358870  0.302603  0.156094   \n",
       "21      0.466788 -0.392280 -0.282131 -0.633253  0.025079 -0.153632  0.263857   \n",
       "22      0.409415  2.814434 -1.269394  1.930274 -0.047710 -0.050892 -0.243216   \n",
       "23      0.333368 -0.039436 -0.700985  0.101164 -0.048362 -0.074431 -0.030156   \n",
       "24     -2.306457 -0.953975 -1.206239  0.686962 -0.186766 -0.698160 -1.664437   \n",
       "25     -0.118283 -0.627893  0.666482 -0.162887 -0.703195 -0.392562  0.311855   \n",
       "26      0.296087 -0.826714 -0.981897 -2.180281  0.329313 -0.035390 -0.122918   \n",
       "27      0.227353 -0.174933 -1.046809  0.528577 -0.145717 -0.116716  0.010199   \n",
       "28      0.048020  1.434696 -0.660974 -0.302093  0.058534 -0.030147  0.354677   \n",
       "29      0.155661  0.384468 -0.375218  0.030722 -0.137716 -0.058033  0.075261   \n",
       "...          ...       ...       ...       ...       ...       ...       ...   \n",
       "149970 -0.223045  0.686651  0.264657 -0.387327 -0.172148  0.017392  0.321762   \n",
       "149971  0.168164  2.785013 -0.086130  0.778481  0.018569  0.188774 -0.486967   \n",
       "149972 -0.338122 -0.537380  4.372295  0.697698 -0.428510  0.611492 -0.424574   \n",
       "149973 -0.060735  0.530556  1.840479  0.199816  1.408987  0.240229  0.321924   \n",
       "149974  0.222540  1.640831  1.254492 -1.287262 -0.637784  0.051539  0.311774   \n",
       "149975 -0.567845 -1.220672  2.953577 -0.010524 -0.445146 -0.783423  0.209066   \n",
       "149976 -0.051097  0.032106  0.825265 -0.263748 -0.541541 -0.219618  0.334055   \n",
       "149977 -0.040305  0.101618 -0.405998  0.281043  0.010296  0.192963  0.302757   \n",
       "149978 -0.293893 -1.078237  0.835051  1.706793 -0.131372 -0.018202 -0.471076   \n",
       "149979  0.832632  2.361111 -2.101367  0.152280 -0.593383  0.037510 -0.573582   \n",
       "149980 -0.184893 -0.999227  0.999803 -0.186210 -0.559111 -0.242803  0.334055   \n",
       "149981 -0.172678 -0.227034  0.482992 -1.278385  0.043991  0.140380  0.237007   \n",
       "149982 -0.210192 -1.177607  1.078993  0.471305 -0.847496 -0.804946  0.323946   \n",
       "149983  0.165646 -1.000763 -0.810031  1.082404  0.152179  0.113169 -0.224009   \n",
       "149984  0.254510 -0.140337  1.753141 -0.855840  0.447327  0.930464 -0.450453   \n",
       "149985 -0.125802 -0.001612  0.454743  0.331630  0.218033  0.127822  0.013555   \n",
       "149986 -0.022101 -0.084473  1.341826 -0.126860  0.129507 -0.052676 -1.173457   \n",
       "149987  0.083941  0.060320  0.617299  0.315958  0.189216  0.154492  0.272227   \n",
       "149988  0.356054  2.294246  0.154244  1.656127  1.730489 -0.338444  0.310642   \n",
       "149989 -0.141664 -0.008585  0.118576  0.721861  0.022403  0.109630 -0.020411   \n",
       "149990  0.566948 -1.845373 -2.557826  0.708178  0.210015  0.187881  0.067942   \n",
       "149991 -0.840501 -0.910830  1.219112  2.135645 -0.031636  0.091237 -0.532013   \n",
       "149992  0.043544 -1.400203 -0.556032 -1.583964  0.352115  0.270728  0.174856   \n",
       "149993 -0.520084 -1.171770  0.392388 -0.610395  0.143045  0.226016  0.310440   \n",
       "149994 -0.132122 -1.254339  0.239633 -0.079566 -0.086058  0.147121  0.116547   \n",
       "149995  0.030564 -0.349722  1.462083  0.469627 -0.878303 -0.798875  0.354273   \n",
       "149996  0.301342 -0.115487  2.360976 -0.549872 -0.711672  1.911911 -0.254902   \n",
       "149997  0.069664  0.705520  1.235442 -0.582444  1.837158  0.510018 -0.079287   \n",
       "149998  0.396852  0.848703 -0.831973  0.858601 -0.507758 -0.172860  0.354233   \n",
       "149999  0.259232 -1.212488 -1.381696 -0.572706  0.263233  0.301439  0.351038   \n",
       "\n",
       "        Fraude  \n",
       "0            0  \n",
       "1            0  \n",
       "2            0  \n",
       "3            0  \n",
       "4            0  \n",
       "5            0  \n",
       "6            0  \n",
       "7            0  \n",
       "8            0  \n",
       "9            0  \n",
       "10           0  \n",
       "11           0  \n",
       "12           0  \n",
       "13           0  \n",
       "14           0  \n",
       "15           0  \n",
       "16           0  \n",
       "17           0  \n",
       "18           0  \n",
       "19           0  \n",
       "20           0  \n",
       "21           0  \n",
       "22           0  \n",
       "23           0  \n",
       "24           0  \n",
       "25           0  \n",
       "26           0  \n",
       "27           0  \n",
       "28           0  \n",
       "29           0  \n",
       "...        ...  \n",
       "149970       0  \n",
       "149971       0  \n",
       "149972       0  \n",
       "149973       0  \n",
       "149974       0  \n",
       "149975       0  \n",
       "149976       0  \n",
       "149977       0  \n",
       "149978       0  \n",
       "149979       0  \n",
       "149980       0  \n",
       "149981       0  \n",
       "149982       0  \n",
       "149983       0  \n",
       "149984       0  \n",
       "149985       0  \n",
       "149986       0  \n",
       "149987       0  \n",
       "149988       0  \n",
       "149989       0  \n",
       "149990       0  \n",
       "149991       0  \n",
       "149992       0  \n",
       "149993       0  \n",
       "149994       0  \n",
       "149995       0  \n",
       "149996       0  \n",
       "149997       0  \n",
       "149998       0  \n",
       "149999       0  \n",
       "\n",
       "[150000 rows x 31 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_zscore = df.copy()\n",
    "\n",
    "df_zscore[features] = df_zscore[features].apply(zscore)\n",
    "\n",
    "df_zscore"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Records with |z| >= 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th># of Outliers</th>\n",
       "      <th>% of Outliers</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PP6</th>\n",
       "      <td>2634</td>\n",
       "      <td>1.756000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP27</th>\n",
       "      <td>2563</td>\n",
       "      <td>1.708667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP20</th>\n",
       "      <td>2527</td>\n",
       "      <td>1.684667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP2</th>\n",
       "      <td>2298</td>\n",
       "      <td>1.532000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sacado</th>\n",
       "      <td>2243</td>\n",
       "      <td>1.495333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP21</th>\n",
       "      <td>2193</td>\n",
       "      <td>1.462000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP8</th>\n",
       "      <td>2174</td>\n",
       "      <td>1.449333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP14</th>\n",
       "      <td>2138</td>\n",
       "      <td>1.425333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP28</th>\n",
       "      <td>2031</td>\n",
       "      <td>1.354000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP12</th>\n",
       "      <td>1906</td>\n",
       "      <td>1.270667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP23</th>\n",
       "      <td>1865</td>\n",
       "      <td>1.243333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP10</th>\n",
       "      <td>1849</td>\n",
       "      <td>1.232667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP7</th>\n",
       "      <td>1840</td>\n",
       "      <td>1.226667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP5</th>\n",
       "      <td>1643</td>\n",
       "      <td>1.095333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP19</th>\n",
       "      <td>1584</td>\n",
       "      <td>1.056000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP17</th>\n",
       "      <td>1559</td>\n",
       "      <td>1.039333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP4</th>\n",
       "      <td>1527</td>\n",
       "      <td>1.018000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP25</th>\n",
       "      <td>1453</td>\n",
       "      <td>0.968667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP9</th>\n",
       "      <td>1156</td>\n",
       "      <td>0.770667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP16</th>\n",
       "      <td>1118</td>\n",
       "      <td>0.745333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP3</th>\n",
       "      <td>1082</td>\n",
       "      <td>0.721333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP18</th>\n",
       "      <td>884</td>\n",
       "      <td>0.589333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP22</th>\n",
       "      <td>729</td>\n",
       "      <td>0.486000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP13</th>\n",
       "      <td>634</td>\n",
       "      <td>0.422667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP15</th>\n",
       "      <td>593</td>\n",
       "      <td>0.395333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP26</th>\n",
       "      <td>470</td>\n",
       "      <td>0.313333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP11</th>\n",
       "      <td>323</td>\n",
       "      <td>0.215333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP24</th>\n",
       "      <td>315</td>\n",
       "      <td>0.210000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP1</th>\n",
       "      <td>7</td>\n",
       "      <td>0.004667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ocorrencia</th>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            # of Outliers  % of Outliers\n",
       "PP6                  2634       1.756000\n",
       "PP27                 2563       1.708667\n",
       "PP20                 2527       1.684667\n",
       "PP2                  2298       1.532000\n",
       "Sacado               2243       1.495333\n",
       "PP21                 2193       1.462000\n",
       "PP8                  2174       1.449333\n",
       "PP14                 2138       1.425333\n",
       "PP28                 2031       1.354000\n",
       "PP12                 1906       1.270667\n",
       "PP23                 1865       1.243333\n",
       "PP10                 1849       1.232667\n",
       "PP7                  1840       1.226667\n",
       "PP5                  1643       1.095333\n",
       "PP19                 1584       1.056000\n",
       "PP17                 1559       1.039333\n",
       "PP4                  1527       1.018000\n",
       "PP25                 1453       0.968667\n",
       "PP9                  1156       0.770667\n",
       "PP16                 1118       0.745333\n",
       "PP3                  1082       0.721333\n",
       "PP18                  884       0.589333\n",
       "PP22                  729       0.486000\n",
       "PP13                  634       0.422667\n",
       "PP15                  593       0.395333\n",
       "PP26                  470       0.313333\n",
       "PP11                  323       0.215333\n",
       "PP24                  315       0.210000\n",
       "PP1                     7       0.004667\n",
       "Ocorrencia              0       0.000000"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "totals = []\n",
    "length = len(df_zscore)\n",
    "\n",
    "for col in df_zscore[features]:\n",
    "    totals.append(df_zscore[col].apply(lambda x: 1 if abs(x) >=3 else 0).sum())\n",
    "    \n",
    "outlier_pct = [(x/length)*100 for x in totals]\n",
    "\n",
    "df_out = pd.DataFrame({'# of Outliers': totals, '% of Outliers': outlier_pct}, index=features)\n",
    "\n",
    "df_out.sort_values(by='% of Outliers', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The variable with the most outliers has ~1.7% of records more than 3 standard deviations away from the mean. Many variables have less than 1% of such cases. Let's check the Fraud distribution for those outliers, in order to find out if they contribute or are just noise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.997771\n",
       "1    0.002229\n",
       "Name: Fraude, dtype: float64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_zscore.loc[abs(df_zscore['Sacado']) >= 3, 'Fraude'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.609907\n",
       "1    0.390093\n",
       "Name: Fraude, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_zscore.loc[abs(df_zscore['PP11']) >= 3, 'Fraude'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.88817\n",
       "1    0.11183\n",
       "Name: Fraude, dtype: float64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_zscore.loc[abs(df_zscore['PP3']) >= 3, 'Fraude'].value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are mixed results. For Sacado the distribution is basically the same as the whole dataframe, while PP3 and PP11 have many outliers that are indeed fraudsters."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Checking Outlier's Fraud Frequencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Fraud Frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PP11</th>\n",
       "      <td>39.009288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP16</th>\n",
       "      <td>12.611807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP17</th>\n",
       "      <td>11.738294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP18</th>\n",
       "      <td>11.199095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP3</th>\n",
       "      <td>11.182994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP14</th>\n",
       "      <td>8.840037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP12</th>\n",
       "      <td>8.761805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP10</th>\n",
       "      <td>8.220660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP4</th>\n",
       "      <td>5.959398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP7</th>\n",
       "      <td>5.869565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP9</th>\n",
       "      <td>5.103806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP5</th>\n",
       "      <td>4.260499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP22</th>\n",
       "      <td>3.155007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP8</th>\n",
       "      <td>2.575897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP2</th>\n",
       "      <td>2.523934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP27</th>\n",
       "      <td>2.301990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP21</th>\n",
       "      <td>2.143183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP19</th>\n",
       "      <td>2.020202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP28</th>\n",
       "      <td>1.280158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP20</th>\n",
       "      <td>1.068461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP15</th>\n",
       "      <td>1.011804</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP23</th>\n",
       "      <td>0.911528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP6</th>\n",
       "      <td>0.797267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP25</th>\n",
       "      <td>0.688231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP13</th>\n",
       "      <td>0.473186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sacado</th>\n",
       "      <td>0.222916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP26</th>\n",
       "      <td>0.212766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP24</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP1</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ocorrencia</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Fraud Frequency\n",
       "PP11              39.009288\n",
       "PP16              12.611807\n",
       "PP17              11.738294\n",
       "PP18              11.199095\n",
       "PP3               11.182994\n",
       "PP14               8.840037\n",
       "PP12               8.761805\n",
       "PP10               8.220660\n",
       "PP4                5.959398\n",
       "PP7                5.869565\n",
       "PP9                5.103806\n",
       "PP5                4.260499\n",
       "PP22               3.155007\n",
       "PP8                2.575897\n",
       "PP2                2.523934\n",
       "PP27               2.301990\n",
       "PP21               2.143183\n",
       "PP19               2.020202\n",
       "PP28               1.280158\n",
       "PP20               1.068461\n",
       "PP15               1.011804\n",
       "PP23               0.911528\n",
       "PP6                0.797267\n",
       "PP25               0.688231\n",
       "PP13               0.473186\n",
       "Sacado             0.222916\n",
       "PP26               0.212766\n",
       "PP24               0.000000\n",
       "PP1                0.000000\n",
       "Ocorrencia         0.000000"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fraud_outlier_frequencies = []\n",
    "\n",
    "for col in df_zscore[features]:\n",
    "    freqs = df_zscore.loc[abs(df_zscore[col]) >= 3, 'Fraude'].value_counts(normalize=True)\n",
    "    if len(freqs) > 1:\n",
    "        freqs = freqs[1]\n",
    "    else:\n",
    "        freqs = 0\n",
    "    fraud_outlier_frequencies.append(freqs*100)\n",
    "        \n",
    "df_out_freq = pd.DataFrame({'Fraud Frequency': fraud_outlier_frequencies}, index=features)\n",
    "\n",
    "df_out_freq.sort_values(by='Fraud Frequency', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Three variables (PP24, PP1 and Ocorrencia) have not even a single Fraudster in their Outliers, meaning it is just noise. In those cases, we will substitute the outliers by the mean."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Substitute Outliers by the Mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "outliers_to_impute = ['PP1', 'PP24']\n",
    "\n",
    "for out in outliers_to_impute:\n",
    "    out_index = list(df_zscore.loc[df_zscore[out] >= 3].index) #indexes to be imputed\n",
    "    mean_val  = df[out].mean() # mean value will substitute outliers\n",
    "    df.loc[out_index, out] = mean_val"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Features with Low Variance <a class=\"anchor\" id=\"6\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, let's take a look at the variance of each feature, to see if there are variables carrying little information. But first all features must be on the same scale - that's why Min Max Scaling will be conducted (bringing everybody to the range [0,1])."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Min Max Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ocorrencia</th>\n",
       "      <th>PP1</th>\n",
       "      <th>PP2</th>\n",
       "      <th>PP3</th>\n",
       "      <th>PP4</th>\n",
       "      <th>PP5</th>\n",
       "      <th>PP6</th>\n",
       "      <th>PP7</th>\n",
       "      <th>PP8</th>\n",
       "      <th>PP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PP20</th>\n",
       "      <th>PP21</th>\n",
       "      <th>PP22</th>\n",
       "      <th>PP23</th>\n",
       "      <th>PP24</th>\n",
       "      <th>PP25</th>\n",
       "      <th>PP26</th>\n",
       "      <th>PP27</th>\n",
       "      <th>PP28</th>\n",
       "      <th>Sacado</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.060108</td>\n",
       "      <td>0.269816</td>\n",
       "      <td>0.194538</td>\n",
       "      <td>0.759211</td>\n",
       "      <td>0.525677</td>\n",
       "      <td>0.454007</td>\n",
       "      <td>0.439219</td>\n",
       "      <td>0.179817</td>\n",
       "      <td>0.536894</td>\n",
       "      <td>...</td>\n",
       "      <td>0.615605</td>\n",
       "      <td>0.444771</td>\n",
       "      <td>0.503320</td>\n",
       "      <td>0.299819</td>\n",
       "      <td>0.908043</td>\n",
       "      <td>0.405191</td>\n",
       "      <td>0.437251</td>\n",
       "      <td>0.348026</td>\n",
       "      <td>0.658603</td>\n",
       "      <td>0.998556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.999989</td>\n",
       "      <td>0.124743</td>\n",
       "      <td>0.280167</td>\n",
       "      <td>0.211427</td>\n",
       "      <td>0.797845</td>\n",
       "      <td>0.535842</td>\n",
       "      <td>0.498879</td>\n",
       "      <td>0.423118</td>\n",
       "      <td>0.185079</td>\n",
       "      <td>0.480048</td>\n",
       "      <td>...</td>\n",
       "      <td>0.598063</td>\n",
       "      <td>0.432270</td>\n",
       "      <td>0.469694</td>\n",
       "      <td>0.305838</td>\n",
       "      <td>0.669805</td>\n",
       "      <td>0.402392</td>\n",
       "      <td>0.540172</td>\n",
       "      <td>0.351041</td>\n",
       "      <td>0.656548</td>\n",
       "      <td>0.979294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.999978</td>\n",
       "      <td>0.437922</td>\n",
       "      <td>0.298739</td>\n",
       "      <td>0.201101</td>\n",
       "      <td>0.729084</td>\n",
       "      <td>0.551182</td>\n",
       "      <td>0.480305</td>\n",
       "      <td>0.424487</td>\n",
       "      <td>0.184995</td>\n",
       "      <td>0.529176</td>\n",
       "      <td>...</td>\n",
       "      <td>0.583944</td>\n",
       "      <td>0.428361</td>\n",
       "      <td>0.471864</td>\n",
       "      <td>0.308728</td>\n",
       "      <td>0.599245</td>\n",
       "      <td>0.418174</td>\n",
       "      <td>0.586579</td>\n",
       "      <td>0.351533</td>\n",
       "      <td>0.654027</td>\n",
       "      <td>0.959301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.999978</td>\n",
       "      <td>0.071959</td>\n",
       "      <td>0.257045</td>\n",
       "      <td>0.212612</td>\n",
       "      <td>0.706865</td>\n",
       "      <td>0.510530</td>\n",
       "      <td>0.485870</td>\n",
       "      <td>0.420289</td>\n",
       "      <td>0.182855</td>\n",
       "      <td>0.545253</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612383</td>\n",
       "      <td>0.436144</td>\n",
       "      <td>0.470058</td>\n",
       "      <td>0.298905</td>\n",
       "      <td>0.737281</td>\n",
       "      <td>0.394715</td>\n",
       "      <td>0.602328</td>\n",
       "      <td>0.349004</td>\n",
       "      <td>0.658670</td>\n",
       "      <td>0.998409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.999966</td>\n",
       "      <td>0.659764</td>\n",
       "      <td>0.288978</td>\n",
       "      <td>0.238128</td>\n",
       "      <td>0.764649</td>\n",
       "      <td>0.534751</td>\n",
       "      <td>0.457382</td>\n",
       "      <td>0.417722</td>\n",
       "      <td>0.170328</td>\n",
       "      <td>0.525486</td>\n",
       "      <td>...</td>\n",
       "      <td>0.621350</td>\n",
       "      <td>0.442086</td>\n",
       "      <td>0.527476</td>\n",
       "      <td>0.318317</td>\n",
       "      <td>0.946516</td>\n",
       "      <td>0.436698</td>\n",
       "      <td>0.404923</td>\n",
       "      <td>0.346270</td>\n",
       "      <td>0.688703</td>\n",
       "      <td>0.973436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.999966</td>\n",
       "      <td>0.060939</td>\n",
       "      <td>0.257608</td>\n",
       "      <td>0.202888</td>\n",
       "      <td>0.694450</td>\n",
       "      <td>0.520002</td>\n",
       "      <td>0.498166</td>\n",
       "      <td>0.423302</td>\n",
       "      <td>0.184329</td>\n",
       "      <td>0.509488</td>\n",
       "      <td>...</td>\n",
       "      <td>0.613390</td>\n",
       "      <td>0.444842</td>\n",
       "      <td>0.537481</td>\n",
       "      <td>0.296726</td>\n",
       "      <td>0.721367</td>\n",
       "      <td>0.399994</td>\n",
       "      <td>0.656492</td>\n",
       "      <td>0.348901</td>\n",
       "      <td>0.658189</td>\n",
       "      <td>0.999885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.999966</td>\n",
       "      <td>0.063501</td>\n",
       "      <td>0.254961</td>\n",
       "      <td>0.214156</td>\n",
       "      <td>0.709450</td>\n",
       "      <td>0.509648</td>\n",
       "      <td>0.492359</td>\n",
       "      <td>0.419326</td>\n",
       "      <td>0.184466</td>\n",
       "      <td>0.547298</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612069</td>\n",
       "      <td>0.437132</td>\n",
       "      <td>0.475351</td>\n",
       "      <td>0.299724</td>\n",
       "      <td>0.738411</td>\n",
       "      <td>0.386322</td>\n",
       "      <td>0.603711</td>\n",
       "      <td>0.349224</td>\n",
       "      <td>0.658782</td>\n",
       "      <td>0.999924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.999966</td>\n",
       "      <td>0.628501</td>\n",
       "      <td>0.314227</td>\n",
       "      <td>0.196994</td>\n",
       "      <td>0.763221</td>\n",
       "      <td>0.470152</td>\n",
       "      <td>0.511660</td>\n",
       "      <td>0.437191</td>\n",
       "      <td>0.179676</td>\n",
       "      <td>0.576492</td>\n",
       "      <td>...</td>\n",
       "      <td>0.589196</td>\n",
       "      <td>0.434704</td>\n",
       "      <td>0.512702</td>\n",
       "      <td>0.275804</td>\n",
       "      <td>0.925919</td>\n",
       "      <td>0.425253</td>\n",
       "      <td>0.494546</td>\n",
       "      <td>0.353528</td>\n",
       "      <td>0.652259</td>\n",
       "      <td>0.980414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.999966</td>\n",
       "      <td>0.058028</td>\n",
       "      <td>0.263326</td>\n",
       "      <td>0.209173</td>\n",
       "      <td>0.811448</td>\n",
       "      <td>0.524605</td>\n",
       "      <td>0.506637</td>\n",
       "      <td>0.424924</td>\n",
       "      <td>0.184904</td>\n",
       "      <td>0.484689</td>\n",
       "      <td>...</td>\n",
       "      <td>0.611292</td>\n",
       "      <td>0.443771</td>\n",
       "      <td>0.523454</td>\n",
       "      <td>0.297067</td>\n",
       "      <td>0.684544</td>\n",
       "      <td>0.401863</td>\n",
       "      <td>0.601175</td>\n",
       "      <td>0.349428</td>\n",
       "      <td>0.658710</td>\n",
       "      <td>0.999942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.999955</td>\n",
       "      <td>0.503509</td>\n",
       "      <td>0.263252</td>\n",
       "      <td>0.182764</td>\n",
       "      <td>0.818657</td>\n",
       "      <td>0.529358</td>\n",
       "      <td>0.504508</td>\n",
       "      <td>0.408831</td>\n",
       "      <td>0.185765</td>\n",
       "      <td>0.594448</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610703</td>\n",
       "      <td>0.438246</td>\n",
       "      <td>0.486868</td>\n",
       "      <td>0.293551</td>\n",
       "      <td>0.594133</td>\n",
       "      <td>0.424509</td>\n",
       "      <td>0.386004</td>\n",
       "      <td>0.351862</td>\n",
       "      <td>0.656021</td>\n",
       "      <td>0.989651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.999944</td>\n",
       "      <td>0.507369</td>\n",
       "      <td>0.249225</td>\n",
       "      <td>0.177989</td>\n",
       "      <td>0.775231</td>\n",
       "      <td>0.509743</td>\n",
       "      <td>0.495201</td>\n",
       "      <td>0.412585</td>\n",
       "      <td>0.184154</td>\n",
       "      <td>0.537584</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612686</td>\n",
       "      <td>0.437830</td>\n",
       "      <td>0.483856</td>\n",
       "      <td>0.303515</td>\n",
       "      <td>0.678457</td>\n",
       "      <td>0.407128</td>\n",
       "      <td>0.490479</td>\n",
       "      <td>0.357300</td>\n",
       "      <td>0.656700</td>\n",
       "      <td>0.999833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.999944</td>\n",
       "      <td>0.068074</td>\n",
       "      <td>0.259608</td>\n",
       "      <td>0.186721</td>\n",
       "      <td>0.694350</td>\n",
       "      <td>0.529656</td>\n",
       "      <td>0.495904</td>\n",
       "      <td>0.429824</td>\n",
       "      <td>0.181994</td>\n",
       "      <td>0.512852</td>\n",
       "      <td>...</td>\n",
       "      <td>0.613391</td>\n",
       "      <td>0.437905</td>\n",
       "      <td>0.482166</td>\n",
       "      <td>0.297311</td>\n",
       "      <td>0.604921</td>\n",
       "      <td>0.406035</td>\n",
       "      <td>0.625691</td>\n",
       "      <td>0.348374</td>\n",
       "      <td>0.658111</td>\n",
       "      <td>0.999940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.999933</td>\n",
       "      <td>0.581976</td>\n",
       "      <td>0.270174</td>\n",
       "      <td>0.166837</td>\n",
       "      <td>0.730940</td>\n",
       "      <td>0.488154</td>\n",
       "      <td>0.472377</td>\n",
       "      <td>0.439546</td>\n",
       "      <td>0.177985</td>\n",
       "      <td>0.507370</td>\n",
       "      <td>...</td>\n",
       "      <td>0.625896</td>\n",
       "      <td>0.440003</td>\n",
       "      <td>0.466206</td>\n",
       "      <td>0.307770</td>\n",
       "      <td>0.837685</td>\n",
       "      <td>0.464918</td>\n",
       "      <td>0.485192</td>\n",
       "      <td>0.344128</td>\n",
       "      <td>0.659633</td>\n",
       "      <td>0.999080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.999921</td>\n",
       "      <td>0.045959</td>\n",
       "      <td>0.265375</td>\n",
       "      <td>0.216054</td>\n",
       "      <td>0.800336</td>\n",
       "      <td>0.518120</td>\n",
       "      <td>0.483458</td>\n",
       "      <td>0.431821</td>\n",
       "      <td>0.183742</td>\n",
       "      <td>0.563240</td>\n",
       "      <td>...</td>\n",
       "      <td>0.607720</td>\n",
       "      <td>0.438863</td>\n",
       "      <td>0.498374</td>\n",
       "      <td>0.300060</td>\n",
       "      <td>0.927711</td>\n",
       "      <td>0.396627</td>\n",
       "      <td>0.600367</td>\n",
       "      <td>0.349539</td>\n",
       "      <td>0.658965</td>\n",
       "      <td>0.999686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.999910</td>\n",
       "      <td>0.055022</td>\n",
       "      <td>0.269408</td>\n",
       "      <td>0.247157</td>\n",
       "      <td>0.826264</td>\n",
       "      <td>0.490317</td>\n",
       "      <td>0.418498</td>\n",
       "      <td>0.432895</td>\n",
       "      <td>0.176874</td>\n",
       "      <td>0.569615</td>\n",
       "      <td>...</td>\n",
       "      <td>0.604352</td>\n",
       "      <td>0.436274</td>\n",
       "      <td>0.489494</td>\n",
       "      <td>0.300122</td>\n",
       "      <td>0.512642</td>\n",
       "      <td>0.386626</td>\n",
       "      <td>0.587311</td>\n",
       "      <td>0.350241</td>\n",
       "      <td>0.658232</td>\n",
       "      <td>0.995587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.999910</td>\n",
       "      <td>0.053504</td>\n",
       "      <td>0.268520</td>\n",
       "      <td>0.197594</td>\n",
       "      <td>0.836429</td>\n",
       "      <td>0.540564</td>\n",
       "      <td>0.522830</td>\n",
       "      <td>0.431134</td>\n",
       "      <td>0.187447</td>\n",
       "      <td>0.522275</td>\n",
       "      <td>...</td>\n",
       "      <td>0.616231</td>\n",
       "      <td>0.443603</td>\n",
       "      <td>0.505097</td>\n",
       "      <td>0.296043</td>\n",
       "      <td>0.565238</td>\n",
       "      <td>0.412659</td>\n",
       "      <td>0.569274</td>\n",
       "      <td>0.348296</td>\n",
       "      <td>0.657760</td>\n",
       "      <td>0.998972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.999910</td>\n",
       "      <td>0.061692</td>\n",
       "      <td>0.257787</td>\n",
       "      <td>0.223630</td>\n",
       "      <td>0.713750</td>\n",
       "      <td>0.509586</td>\n",
       "      <td>0.492926</td>\n",
       "      <td>0.419698</td>\n",
       "      <td>0.184235</td>\n",
       "      <td>0.534748</td>\n",
       "      <td>...</td>\n",
       "      <td>0.611983</td>\n",
       "      <td>0.438476</td>\n",
       "      <td>0.496585</td>\n",
       "      <td>0.301951</td>\n",
       "      <td>0.838281</td>\n",
       "      <td>0.380579</td>\n",
       "      <td>0.599394</td>\n",
       "      <td>0.350318</td>\n",
       "      <td>0.658867</td>\n",
       "      <td>0.998008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.999899</td>\n",
       "      <td>0.491944</td>\n",
       "      <td>0.247020</td>\n",
       "      <td>0.210613</td>\n",
       "      <td>0.662371</td>\n",
       "      <td>0.485535</td>\n",
       "      <td>0.384332</td>\n",
       "      <td>0.430031</td>\n",
       "      <td>0.168089</td>\n",
       "      <td>0.572563</td>\n",
       "      <td>...</td>\n",
       "      <td>0.607939</td>\n",
       "      <td>0.435439</td>\n",
       "      <td>0.469519</td>\n",
       "      <td>0.300365</td>\n",
       "      <td>0.510056</td>\n",
       "      <td>0.432115</td>\n",
       "      <td>0.495633</td>\n",
       "      <td>0.351272</td>\n",
       "      <td>0.656911</td>\n",
       "      <td>0.997636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.999899</td>\n",
       "      <td>0.068406</td>\n",
       "      <td>0.259734</td>\n",
       "      <td>0.206104</td>\n",
       "      <td>0.714286</td>\n",
       "      <td>0.521034</td>\n",
       "      <td>0.501808</td>\n",
       "      <td>0.421213</td>\n",
       "      <td>0.184896</td>\n",
       "      <td>0.528196</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610687</td>\n",
       "      <td>0.441122</td>\n",
       "      <td>0.509858</td>\n",
       "      <td>0.298266</td>\n",
       "      <td>0.628529</td>\n",
       "      <td>0.399882</td>\n",
       "      <td>0.510058</td>\n",
       "      <td>0.351072</td>\n",
       "      <td>0.658633</td>\n",
       "      <td>0.997664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.999899</td>\n",
       "      <td>0.091646</td>\n",
       "      <td>0.263529</td>\n",
       "      <td>0.212942</td>\n",
       "      <td>0.687926</td>\n",
       "      <td>0.514817</td>\n",
       "      <td>0.478196</td>\n",
       "      <td>0.420261</td>\n",
       "      <td>0.181587</td>\n",
       "      <td>0.529101</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610140</td>\n",
       "      <td>0.437772</td>\n",
       "      <td>0.491034</td>\n",
       "      <td>0.300148</td>\n",
       "      <td>0.740123</td>\n",
       "      <td>0.396474</td>\n",
       "      <td>0.606181</td>\n",
       "      <td>0.349557</td>\n",
       "      <td>0.658190</td>\n",
       "      <td>0.992420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.999888</td>\n",
       "      <td>0.603368</td>\n",
       "      <td>0.245626</td>\n",
       "      <td>0.193507</td>\n",
       "      <td>0.778744</td>\n",
       "      <td>0.538936</td>\n",
       "      <td>0.508206</td>\n",
       "      <td>0.430635</td>\n",
       "      <td>0.169207</td>\n",
       "      <td>0.528025</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609587</td>\n",
       "      <td>0.436557</td>\n",
       "      <td>0.478335</td>\n",
       "      <td>0.301043</td>\n",
       "      <td>0.606673</td>\n",
       "      <td>0.397660</td>\n",
       "      <td>0.370018</td>\n",
       "      <td>0.345905</td>\n",
       "      <td>0.661565</td>\n",
       "      <td>0.997456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0.999876</td>\n",
       "      <td>0.547288</td>\n",
       "      <td>0.263441</td>\n",
       "      <td>0.146722</td>\n",
       "      <td>0.742337</td>\n",
       "      <td>0.516630</td>\n",
       "      <td>0.478693</td>\n",
       "      <td>0.429831</td>\n",
       "      <td>0.177433</td>\n",
       "      <td>0.497740</td>\n",
       "      <td>...</td>\n",
       "      <td>0.615299</td>\n",
       "      <td>0.440121</td>\n",
       "      <td>0.490750</td>\n",
       "      <td>0.302470</td>\n",
       "      <td>0.647573</td>\n",
       "      <td>0.412102</td>\n",
       "      <td>0.500056</td>\n",
       "      <td>0.350303</td>\n",
       "      <td>0.657476</td>\n",
       "      <td>0.998812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.999876</td>\n",
       "      <td>0.070490</td>\n",
       "      <td>0.263707</td>\n",
       "      <td>0.243113</td>\n",
       "      <td>0.813920</td>\n",
       "      <td>0.501833</td>\n",
       "      <td>0.488295</td>\n",
       "      <td>0.413711</td>\n",
       "      <td>0.186118</td>\n",
       "      <td>0.515571</td>\n",
       "      <td>...</td>\n",
       "      <td>0.605819</td>\n",
       "      <td>0.443106</td>\n",
       "      <td>0.537544</td>\n",
       "      <td>0.301911</td>\n",
       "      <td>0.980834</td>\n",
       "      <td>0.384053</td>\n",
       "      <td>0.712976</td>\n",
       "      <td>0.349469</td>\n",
       "      <td>0.658396</td>\n",
       "      <td>0.992433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.999865</td>\n",
       "      <td>0.064571</td>\n",
       "      <td>0.272311</td>\n",
       "      <td>0.196872</td>\n",
       "      <td>0.774924</td>\n",
       "      <td>0.537216</td>\n",
       "      <td>0.492495</td>\n",
       "      <td>0.436480</td>\n",
       "      <td>0.182498</td>\n",
       "      <td>0.541370</td>\n",
       "      <td>...</td>\n",
       "      <td>0.607089</td>\n",
       "      <td>0.432516</td>\n",
       "      <td>0.450391</td>\n",
       "      <td>0.301169</td>\n",
       "      <td>0.684242</td>\n",
       "      <td>0.400202</td>\n",
       "      <td>0.561055</td>\n",
       "      <td>0.349462</td>\n",
       "      <td>0.658185</td>\n",
       "      <td>0.995113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0.999854</td>\n",
       "      <td>0.622654</td>\n",
       "      <td>0.302666</td>\n",
       "      <td>0.166313</td>\n",
       "      <td>0.770253</td>\n",
       "      <td>0.514497</td>\n",
       "      <td>0.513362</td>\n",
       "      <td>0.423665</td>\n",
       "      <td>0.181601</td>\n",
       "      <td>0.556268</td>\n",
       "      <td>...</td>\n",
       "      <td>0.580306</td>\n",
       "      <td>0.429427</td>\n",
       "      <td>0.476694</td>\n",
       "      <td>0.275411</td>\n",
       "      <td>0.589198</td>\n",
       "      <td>0.385847</td>\n",
       "      <td>0.609710</td>\n",
       "      <td>0.347876</td>\n",
       "      <td>0.652595</td>\n",
       "      <td>0.974552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0.999843</td>\n",
       "      <td>0.468042</td>\n",
       "      <td>0.246626</td>\n",
       "      <td>0.179216</td>\n",
       "      <td>0.748115</td>\n",
       "      <td>0.512977</td>\n",
       "      <td>0.506617</td>\n",
       "      <td>0.410903</td>\n",
       "      <td>0.184700</td>\n",
       "      <td>0.552606</td>\n",
       "      <td>...</td>\n",
       "      <td>0.608547</td>\n",
       "      <td>0.441297</td>\n",
       "      <td>0.509048</td>\n",
       "      <td>0.296762</td>\n",
       "      <td>0.623086</td>\n",
       "      <td>0.439052</td>\n",
       "      <td>0.539124</td>\n",
       "      <td>0.341961</td>\n",
       "      <td>0.655334</td>\n",
       "      <td>0.999416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.999843</td>\n",
       "      <td>0.068820</td>\n",
       "      <td>0.259389</td>\n",
       "      <td>0.225439</td>\n",
       "      <td>0.740207</td>\n",
       "      <td>0.515899</td>\n",
       "      <td>0.524262</td>\n",
       "      <td>0.411517</td>\n",
       "      <td>0.190307</td>\n",
       "      <td>0.546224</td>\n",
       "      <td>...</td>\n",
       "      <td>0.606831</td>\n",
       "      <td>0.438082</td>\n",
       "      <td>0.495486</td>\n",
       "      <td>0.300805</td>\n",
       "      <td>0.602424</td>\n",
       "      <td>0.392221</td>\n",
       "      <td>0.371565</td>\n",
       "      <td>0.353787</td>\n",
       "      <td>0.658535</td>\n",
       "      <td>0.993946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>0.999843</td>\n",
       "      <td>0.078209</td>\n",
       "      <td>0.261978</td>\n",
       "      <td>0.198220</td>\n",
       "      <td>0.683821</td>\n",
       "      <td>0.522798</td>\n",
       "      <td>0.487356</td>\n",
       "      <td>0.424311</td>\n",
       "      <td>0.183387</td>\n",
       "      <td>0.506036</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610572</td>\n",
       "      <td>0.438074</td>\n",
       "      <td>0.479237</td>\n",
       "      <td>0.300134</td>\n",
       "      <td>0.670161</td>\n",
       "      <td>0.390377</td>\n",
       "      <td>0.596555</td>\n",
       "      <td>0.348346</td>\n",
       "      <td>0.657806</td>\n",
       "      <td>0.995621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>0.999831</td>\n",
       "      <td>0.056515</td>\n",
       "      <td>0.254350</td>\n",
       "      <td>0.221146</td>\n",
       "      <td>0.734918</td>\n",
       "      <td>0.507038</td>\n",
       "      <td>0.494405</td>\n",
       "      <td>0.419691</td>\n",
       "      <td>0.185488</td>\n",
       "      <td>0.546602</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609531</td>\n",
       "      <td>0.443414</td>\n",
       "      <td>0.530226</td>\n",
       "      <td>0.298384</td>\n",
       "      <td>0.837443</td>\n",
       "      <td>0.401339</td>\n",
       "      <td>0.527562</td>\n",
       "      <td>0.350686</td>\n",
       "      <td>0.658582</td>\n",
       "      <td>0.999955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0.999831</td>\n",
       "      <td>0.067138</td>\n",
       "      <td>0.269948</td>\n",
       "      <td>0.197098</td>\n",
       "      <td>0.783340</td>\n",
       "      <td>0.529933</td>\n",
       "      <td>0.479951</td>\n",
       "      <td>0.436058</td>\n",
       "      <td>0.181232</td>\n",
       "      <td>0.559892</td>\n",
       "      <td>...</td>\n",
       "      <td>0.607504</td>\n",
       "      <td>0.430085</td>\n",
       "      <td>0.426398</td>\n",
       "      <td>0.299435</td>\n",
       "      <td>0.728297</td>\n",
       "      <td>0.409457</td>\n",
       "      <td>0.555204</td>\n",
       "      <td>0.348438</td>\n",
       "      <td>0.658332</td>\n",
       "      <td>0.996439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149970</th>\n",
       "      <td>0.000157</td>\n",
       "      <td>0.532614</td>\n",
       "      <td>0.233550</td>\n",
       "      <td>0.267199</td>\n",
       "      <td>0.781760</td>\n",
       "      <td>0.505738</td>\n",
       "      <td>0.503238</td>\n",
       "      <td>0.421217</td>\n",
       "      <td>0.171538</td>\n",
       "      <td>0.543447</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609687</td>\n",
       "      <td>0.443950</td>\n",
       "      <td>0.539304</td>\n",
       "      <td>0.295740</td>\n",
       "      <td>0.759702</td>\n",
       "      <td>0.427636</td>\n",
       "      <td>0.520482</td>\n",
       "      <td>0.348044</td>\n",
       "      <td>0.659008</td>\n",
       "      <td>0.999541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149971</th>\n",
       "      <td>0.000157</td>\n",
       "      <td>0.046490</td>\n",
       "      <td>0.264995</td>\n",
       "      <td>0.270647</td>\n",
       "      <td>0.702956</td>\n",
       "      <td>0.491346</td>\n",
       "      <td>0.467347</td>\n",
       "      <td>0.411664</td>\n",
       "      <td>0.182871</td>\n",
       "      <td>0.549368</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610117</td>\n",
       "      <td>0.432241</td>\n",
       "      <td>0.449527</td>\n",
       "      <td>0.299557</td>\n",
       "      <td>0.977777</td>\n",
       "      <td>0.417670</td>\n",
       "      <td>0.617311</td>\n",
       "      <td>0.350228</td>\n",
       "      <td>0.660544</td>\n",
       "      <td>0.989366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149972</th>\n",
       "      <td>0.000157</td>\n",
       "      <td>0.481101</td>\n",
       "      <td>0.265155</td>\n",
       "      <td>0.245330</td>\n",
       "      <td>0.747723</td>\n",
       "      <td>0.507301</td>\n",
       "      <td>0.490690</td>\n",
       "      <td>0.411432</td>\n",
       "      <td>0.183605</td>\n",
       "      <td>0.521229</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612440</td>\n",
       "      <td>0.436974</td>\n",
       "      <td>0.469025</td>\n",
       "      <td>0.294617</td>\n",
       "      <td>0.632493</td>\n",
       "      <td>0.544335</td>\n",
       "      <td>0.610601</td>\n",
       "      <td>0.345107</td>\n",
       "      <td>0.664333</td>\n",
       "      <td>0.990151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149973</th>\n",
       "      <td>0.000146</td>\n",
       "      <td>0.508196</td>\n",
       "      <td>0.247321</td>\n",
       "      <td>0.217955</td>\n",
       "      <td>0.798762</td>\n",
       "      <td>0.500127</td>\n",
       "      <td>0.510187</td>\n",
       "      <td>0.407694</td>\n",
       "      <td>0.185739</td>\n",
       "      <td>0.548191</td>\n",
       "      <td>...</td>\n",
       "      <td>0.614358</td>\n",
       "      <td>0.439940</td>\n",
       "      <td>0.503672</td>\n",
       "      <td>0.297323</td>\n",
       "      <td>0.743479</td>\n",
       "      <td>0.472405</td>\n",
       "      <td>0.569249</td>\n",
       "      <td>0.366154</td>\n",
       "      <td>0.661006</td>\n",
       "      <td>0.999543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149974</th>\n",
       "      <td>0.000146</td>\n",
       "      <td>0.470533</td>\n",
       "      <td>0.254848</td>\n",
       "      <td>0.180369</td>\n",
       "      <td>0.783546</td>\n",
       "      <td>0.503923</td>\n",
       "      <td>0.467158</td>\n",
       "      <td>0.417149</td>\n",
       "      <td>0.181294</td>\n",
       "      <td>0.513165</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610052</td>\n",
       "      <td>0.434837</td>\n",
       "      <td>0.435652</td>\n",
       "      <td>0.300087</td>\n",
       "      <td>0.858866</td>\n",
       "      <td>0.455757</td>\n",
       "      <td>0.445736</td>\n",
       "      <td>0.342710</td>\n",
       "      <td>0.659314</td>\n",
       "      <td>0.999415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149975</th>\n",
       "      <td>0.000135</td>\n",
       "      <td>0.409908</td>\n",
       "      <td>0.253989</td>\n",
       "      <td>0.195949</td>\n",
       "      <td>0.788477</td>\n",
       "      <td>0.509173</td>\n",
       "      <td>0.492489</td>\n",
       "      <td>0.414502</td>\n",
       "      <td>0.183078</td>\n",
       "      <td>0.536575</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610661</td>\n",
       "      <td>0.439295</td>\n",
       "      <td>0.500524</td>\n",
       "      <td>0.292375</td>\n",
       "      <td>0.561481</td>\n",
       "      <td>0.504029</td>\n",
       "      <td>0.551778</td>\n",
       "      <td>0.344917</td>\n",
       "      <td>0.651831</td>\n",
       "      <td>0.998123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149976</th>\n",
       "      <td>0.000124</td>\n",
       "      <td>0.182477</td>\n",
       "      <td>0.247111</td>\n",
       "      <td>0.232192</td>\n",
       "      <td>0.782715</td>\n",
       "      <td>0.493247</td>\n",
       "      <td>0.497517</td>\n",
       "      <td>0.408425</td>\n",
       "      <td>0.184280</td>\n",
       "      <td>0.545814</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609134</td>\n",
       "      <td>0.443769</td>\n",
       "      <td>0.528883</td>\n",
       "      <td>0.297417</td>\n",
       "      <td>0.691677</td>\n",
       "      <td>0.443563</td>\n",
       "      <td>0.530746</td>\n",
       "      <td>0.343813</td>\n",
       "      <td>0.656884</td>\n",
       "      <td>0.999695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149977</th>\n",
       "      <td>0.000124</td>\n",
       "      <td>0.019006</td>\n",
       "      <td>0.261519</td>\n",
       "      <td>0.243387</td>\n",
       "      <td>0.722387</td>\n",
       "      <td>0.515834</td>\n",
       "      <td>0.513095</td>\n",
       "      <td>0.419868</td>\n",
       "      <td>0.187521</td>\n",
       "      <td>0.495593</td>\n",
       "      <td>...</td>\n",
       "      <td>0.614397</td>\n",
       "      <td>0.437045</td>\n",
       "      <td>0.464115</td>\n",
       "      <td>0.297523</td>\n",
       "      <td>0.698902</td>\n",
       "      <td>0.408582</td>\n",
       "      <td>0.575995</td>\n",
       "      <td>0.350133</td>\n",
       "      <td>0.660582</td>\n",
       "      <td>0.999302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149978</th>\n",
       "      <td>0.000112</td>\n",
       "      <td>0.042613</td>\n",
       "      <td>0.271372</td>\n",
       "      <td>0.255537</td>\n",
       "      <td>0.730603</td>\n",
       "      <td>0.487255</td>\n",
       "      <td>0.395535</td>\n",
       "      <td>0.433196</td>\n",
       "      <td>0.173116</td>\n",
       "      <td>0.506258</td>\n",
       "      <td>...</td>\n",
       "      <td>0.606196</td>\n",
       "      <td>0.437104</td>\n",
       "      <td>0.497376</td>\n",
       "      <td>0.295048</td>\n",
       "      <td>0.576284</td>\n",
       "      <td>0.443841</td>\n",
       "      <td>0.694414</td>\n",
       "      <td>0.348511</td>\n",
       "      <td>0.658689</td>\n",
       "      <td>0.989566</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149979</th>\n",
       "      <td>0.000112</td>\n",
       "      <td>0.571725</td>\n",
       "      <td>0.276648</td>\n",
       "      <td>0.218708</td>\n",
       "      <td>0.835705</td>\n",
       "      <td>0.489958</td>\n",
       "      <td>0.506151</td>\n",
       "      <td>0.419653</td>\n",
       "      <td>0.185217</td>\n",
       "      <td>0.571860</td>\n",
       "      <td>...</td>\n",
       "      <td>0.606836</td>\n",
       "      <td>0.439324</td>\n",
       "      <td>0.494412</td>\n",
       "      <td>0.306040</td>\n",
       "      <td>0.933722</td>\n",
       "      <td>0.360417</td>\n",
       "      <td>0.565301</td>\n",
       "      <td>0.343219</td>\n",
       "      <td>0.659189</td>\n",
       "      <td>0.988276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149980</th>\n",
       "      <td>0.000112</td>\n",
       "      <td>0.200347</td>\n",
       "      <td>0.246626</td>\n",
       "      <td>0.225561</td>\n",
       "      <td>0.776334</td>\n",
       "      <td>0.497161</td>\n",
       "      <td>0.504085</td>\n",
       "      <td>0.408024</td>\n",
       "      <td>0.184750</td>\n",
       "      <td>0.549782</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609117</td>\n",
       "      <td>0.442888</td>\n",
       "      <td>0.520362</td>\n",
       "      <td>0.296112</td>\n",
       "      <td>0.584495</td>\n",
       "      <td>0.448522</td>\n",
       "      <td>0.537186</td>\n",
       "      <td>0.343611</td>\n",
       "      <td>0.656676</td>\n",
       "      <td>0.999695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149981</th>\n",
       "      <td>0.000101</td>\n",
       "      <td>0.021856</td>\n",
       "      <td>0.263580</td>\n",
       "      <td>0.222621</td>\n",
       "      <td>0.725007</td>\n",
       "      <td>0.524848</td>\n",
       "      <td>0.506250</td>\n",
       "      <td>0.427495</td>\n",
       "      <td>0.186490</td>\n",
       "      <td>0.498180</td>\n",
       "      <td>...</td>\n",
       "      <td>0.611260</td>\n",
       "      <td>0.434648</td>\n",
       "      <td>0.446650</td>\n",
       "      <td>0.296231</td>\n",
       "      <td>0.664746</td>\n",
       "      <td>0.433839</td>\n",
       "      <td>0.446474</td>\n",
       "      <td>0.350519</td>\n",
       "      <td>0.660111</td>\n",
       "      <td>0.998474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149982</th>\n",
       "      <td>0.000090</td>\n",
       "      <td>0.567706</td>\n",
       "      <td>0.229124</td>\n",
       "      <td>0.265600</td>\n",
       "      <td>0.770859</td>\n",
       "      <td>0.519726</td>\n",
       "      <td>0.515122</td>\n",
       "      <td>0.425296</td>\n",
       "      <td>0.168247</td>\n",
       "      <td>0.545042</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610050</td>\n",
       "      <td>0.431271</td>\n",
       "      <td>0.432195</td>\n",
       "      <td>0.295865</td>\n",
       "      <td>0.565957</td>\n",
       "      <td>0.450771</td>\n",
       "      <td>0.591798</td>\n",
       "      <td>0.340308</td>\n",
       "      <td>0.651638</td>\n",
       "      <td>0.999568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149983</th>\n",
       "      <td>0.000079</td>\n",
       "      <td>0.028427</td>\n",
       "      <td>0.259189</td>\n",
       "      <td>0.268290</td>\n",
       "      <td>0.694166</td>\n",
       "      <td>0.499947</td>\n",
       "      <td>0.513015</td>\n",
       "      <td>0.407354</td>\n",
       "      <td>0.190131</td>\n",
       "      <td>0.543477</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609372</td>\n",
       "      <td>0.435329</td>\n",
       "      <td>0.474033</td>\n",
       "      <td>0.299532</td>\n",
       "      <td>0.584335</td>\n",
       "      <td>0.397104</td>\n",
       "      <td>0.642554</td>\n",
       "      <td>0.351759</td>\n",
       "      <td>0.659867</td>\n",
       "      <td>0.992674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149984</th>\n",
       "      <td>0.000067</td>\n",
       "      <td>0.186562</td>\n",
       "      <td>0.254587</td>\n",
       "      <td>0.198728</td>\n",
       "      <td>0.723938</td>\n",
       "      <td>0.518759</td>\n",
       "      <td>0.460384</td>\n",
       "      <td>0.405285</td>\n",
       "      <td>0.191462</td>\n",
       "      <td>0.514548</td>\n",
       "      <td>...</td>\n",
       "      <td>0.604728</td>\n",
       "      <td>0.436449</td>\n",
       "      <td>0.435638</td>\n",
       "      <td>0.300399</td>\n",
       "      <td>0.673756</td>\n",
       "      <td>0.469924</td>\n",
       "      <td>0.481569</td>\n",
       "      <td>0.355139</td>\n",
       "      <td>0.667192</td>\n",
       "      <td>0.989825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149985</th>\n",
       "      <td>0.000067</td>\n",
       "      <td>0.202279</td>\n",
       "      <td>0.246037</td>\n",
       "      <td>0.242916</td>\n",
       "      <td>0.746867</td>\n",
       "      <td>0.509363</td>\n",
       "      <td>0.522950</td>\n",
       "      <td>0.407422</td>\n",
       "      <td>0.185101</td>\n",
       "      <td>0.550593</td>\n",
       "      <td>...</td>\n",
       "      <td>0.609743</td>\n",
       "      <td>0.432266</td>\n",
       "      <td>0.440902</td>\n",
       "      <td>0.296688</td>\n",
       "      <td>0.688173</td>\n",
       "      <td>0.433036</td>\n",
       "      <td>0.580197</td>\n",
       "      <td>0.352513</td>\n",
       "      <td>0.659998</td>\n",
       "      <td>0.995663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149986</th>\n",
       "      <td>0.000067</td>\n",
       "      <td>0.065040</td>\n",
       "      <td>0.279523</td>\n",
       "      <td>0.221239</td>\n",
       "      <td>0.700656</td>\n",
       "      <td>0.533802</td>\n",
       "      <td>0.485056</td>\n",
       "      <td>0.426519</td>\n",
       "      <td>0.183214</td>\n",
       "      <td>0.487101</td>\n",
       "      <td>...</td>\n",
       "      <td>0.601667</td>\n",
       "      <td>0.437088</td>\n",
       "      <td>0.501360</td>\n",
       "      <td>0.297700</td>\n",
       "      <td>0.679562</td>\n",
       "      <td>0.458239</td>\n",
       "      <td>0.542116</td>\n",
       "      <td>0.351499</td>\n",
       "      <td>0.658380</td>\n",
       "      <td>0.980729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149987</th>\n",
       "      <td>0.000067</td>\n",
       "      <td>0.171513</td>\n",
       "      <td>0.245399</td>\n",
       "      <td>0.241619</td>\n",
       "      <td>0.747931</td>\n",
       "      <td>0.508853</td>\n",
       "      <td>0.526252</td>\n",
       "      <td>0.411864</td>\n",
       "      <td>0.183825</td>\n",
       "      <td>0.541090</td>\n",
       "      <td>...</td>\n",
       "      <td>0.613549</td>\n",
       "      <td>0.433180</td>\n",
       "      <td>0.447016</td>\n",
       "      <td>0.298735</td>\n",
       "      <td>0.694610</td>\n",
       "      <td>0.437655</td>\n",
       "      <td>0.578895</td>\n",
       "      <td>0.352183</td>\n",
       "      <td>0.660237</td>\n",
       "      <td>0.998917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149988</th>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.533147</td>\n",
       "      <td>0.241624</td>\n",
       "      <td>0.228711</td>\n",
       "      <td>0.827618</td>\n",
       "      <td>0.493825</td>\n",
       "      <td>0.494891</td>\n",
       "      <td>0.406833</td>\n",
       "      <td>0.183603</td>\n",
       "      <td>0.541563</td>\n",
       "      <td>...</td>\n",
       "      <td>0.615864</td>\n",
       "      <td>0.439238</td>\n",
       "      <td>0.508968</td>\n",
       "      <td>0.301390</td>\n",
       "      <td>0.926773</td>\n",
       "      <td>0.424499</td>\n",
       "      <td>0.690206</td>\n",
       "      <td>0.369836</td>\n",
       "      <td>0.655819</td>\n",
       "      <td>0.999401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149989</th>\n",
       "      <td>0.000056</td>\n",
       "      <td>0.026445</td>\n",
       "      <td>0.262856</td>\n",
       "      <td>0.241106</td>\n",
       "      <td>0.724946</td>\n",
       "      <td>0.516910</td>\n",
       "      <td>0.514813</td>\n",
       "      <td>0.418072</td>\n",
       "      <td>0.188466</td>\n",
       "      <td>0.507721</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610454</td>\n",
       "      <td>0.437349</td>\n",
       "      <td>0.478786</td>\n",
       "      <td>0.296534</td>\n",
       "      <td>0.687449</td>\n",
       "      <td>0.423486</td>\n",
       "      <td>0.612608</td>\n",
       "      <td>0.350272</td>\n",
       "      <td>0.659835</td>\n",
       "      <td>0.995236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149990</th>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.462767</td>\n",
       "      <td>0.245008</td>\n",
       "      <td>0.206387</td>\n",
       "      <td>0.721055</td>\n",
       "      <td>0.496705</td>\n",
       "      <td>0.503385</td>\n",
       "      <td>0.396524</td>\n",
       "      <td>0.188572</td>\n",
       "      <td>0.575909</td>\n",
       "      <td>...</td>\n",
       "      <td>0.607776</td>\n",
       "      <td>0.436937</td>\n",
       "      <td>0.475566</td>\n",
       "      <td>0.303448</td>\n",
       "      <td>0.496558</td>\n",
       "      <td>0.347448</td>\n",
       "      <td>0.611472</td>\n",
       "      <td>0.352421</td>\n",
       "      <td>0.660536</td>\n",
       "      <td>0.996347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149991</th>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.491811</td>\n",
       "      <td>0.252618</td>\n",
       "      <td>0.212195</td>\n",
       "      <td>0.690086</td>\n",
       "      <td>0.520855</td>\n",
       "      <td>0.472870</td>\n",
       "      <td>0.408565</td>\n",
       "      <td>0.180919</td>\n",
       "      <td>0.535867</td>\n",
       "      <td>...</td>\n",
       "      <td>0.611110</td>\n",
       "      <td>0.444796</td>\n",
       "      <td>0.535423</td>\n",
       "      <td>0.289715</td>\n",
       "      <td>0.593682</td>\n",
       "      <td>0.454752</td>\n",
       "      <td>0.730033</td>\n",
       "      <td>0.349653</td>\n",
       "      <td>0.659670</td>\n",
       "      <td>0.988799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149992</th>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.019255</td>\n",
       "      <td>0.259797</td>\n",
       "      <td>0.266591</td>\n",
       "      <td>0.741629</td>\n",
       "      <td>0.502813</td>\n",
       "      <td>0.513255</td>\n",
       "      <td>0.411977</td>\n",
       "      <td>0.188803</td>\n",
       "      <td>0.540003</td>\n",
       "      <td>...</td>\n",
       "      <td>0.611673</td>\n",
       "      <td>0.435822</td>\n",
       "      <td>0.466025</td>\n",
       "      <td>0.298341</td>\n",
       "      <td>0.542823</td>\n",
       "      <td>0.404320</td>\n",
       "      <td>0.421093</td>\n",
       "      <td>0.354049</td>\n",
       "      <td>0.661279</td>\n",
       "      <td>0.997692</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149993</th>\n",
       "      <td>0.000034</td>\n",
       "      <td>0.018064</td>\n",
       "      <td>0.263096</td>\n",
       "      <td>0.278815</td>\n",
       "      <td>0.778208</td>\n",
       "      <td>0.474551</td>\n",
       "      <td>0.415605</td>\n",
       "      <td>0.425636</td>\n",
       "      <td>0.176473</td>\n",
       "      <td>0.522033</td>\n",
       "      <td>...</td>\n",
       "      <td>0.611982</td>\n",
       "      <td>0.442510</td>\n",
       "      <td>0.521803</td>\n",
       "      <td>0.292841</td>\n",
       "      <td>0.566563</td>\n",
       "      <td>0.431265</td>\n",
       "      <td>0.501955</td>\n",
       "      <td>0.351654</td>\n",
       "      <td>0.660878</td>\n",
       "      <td>0.999398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149994</th>\n",
       "      <td>0.000034</td>\n",
       "      <td>0.017594</td>\n",
       "      <td>0.276054</td>\n",
       "      <td>0.258699</td>\n",
       "      <td>0.828557</td>\n",
       "      <td>0.495146</td>\n",
       "      <td>0.403561</td>\n",
       "      <td>0.443447</td>\n",
       "      <td>0.173894</td>\n",
       "      <td>0.530968</td>\n",
       "      <td>...</td>\n",
       "      <td>0.607534</td>\n",
       "      <td>0.431134</td>\n",
       "      <td>0.431869</td>\n",
       "      <td>0.296627</td>\n",
       "      <td>0.557982</td>\n",
       "      <td>0.426925</td>\n",
       "      <td>0.546044</td>\n",
       "      <td>0.349030</td>\n",
       "      <td>0.660171</td>\n",
       "      <td>0.996959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149995</th>\n",
       "      <td>0.000011</td>\n",
       "      <td>0.497450</td>\n",
       "      <td>0.241053</td>\n",
       "      <td>0.247879</td>\n",
       "      <td>0.778489</td>\n",
       "      <td>0.507282</td>\n",
       "      <td>0.500286</td>\n",
       "      <td>0.419056</td>\n",
       "      <td>0.174705</td>\n",
       "      <td>0.543771</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612283</td>\n",
       "      <td>0.431614</td>\n",
       "      <td>0.435018</td>\n",
       "      <td>0.298214</td>\n",
       "      <td>0.651996</td>\n",
       "      <td>0.461655</td>\n",
       "      <td>0.591658</td>\n",
       "      <td>0.339955</td>\n",
       "      <td>0.651693</td>\n",
       "      <td>0.999950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149996</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.390792</td>\n",
       "      <td>0.235442</td>\n",
       "      <td>0.204946</td>\n",
       "      <td>0.569748</td>\n",
       "      <td>0.504264</td>\n",
       "      <td>0.480996</td>\n",
       "      <td>0.387067</td>\n",
       "      <td>0.199926</td>\n",
       "      <td>0.556866</td>\n",
       "      <td>...</td>\n",
       "      <td>0.594849</td>\n",
       "      <td>0.439691</td>\n",
       "      <td>0.425836</td>\n",
       "      <td>0.300856</td>\n",
       "      <td>0.676339</td>\n",
       "      <td>0.487193</td>\n",
       "      <td>0.506982</td>\n",
       "      <td>0.341864</td>\n",
       "      <td>0.675988</td>\n",
       "      <td>0.992286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149997</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.568302</td>\n",
       "      <td>0.247759</td>\n",
       "      <td>0.227228</td>\n",
       "      <td>0.756020</td>\n",
       "      <td>0.507406</td>\n",
       "      <td>0.495916</td>\n",
       "      <td>0.402733</td>\n",
       "      <td>0.190475</td>\n",
       "      <td>0.526601</td>\n",
       "      <td>...</td>\n",
       "      <td>0.614769</td>\n",
       "      <td>0.444031</td>\n",
       "      <td>0.497298</td>\n",
       "      <td>0.298596</td>\n",
       "      <td>0.761663</td>\n",
       "      <td>0.455216</td>\n",
       "      <td>0.504276</td>\n",
       "      <td>0.371058</td>\n",
       "      <td>0.663423</td>\n",
       "      <td>0.994495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149998</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.521388</td>\n",
       "      <td>0.247262</td>\n",
       "      <td>0.179510</td>\n",
       "      <td>0.764294</td>\n",
       "      <td>0.516755</td>\n",
       "      <td>0.481496</td>\n",
       "      <td>0.421304</td>\n",
       "      <td>0.175882</td>\n",
       "      <td>0.525186</td>\n",
       "      <td>...</td>\n",
       "      <td>0.610705</td>\n",
       "      <td>0.444292</td>\n",
       "      <td>0.529266</td>\n",
       "      <td>0.301788</td>\n",
       "      <td>0.776543</td>\n",
       "      <td>0.396480</td>\n",
       "      <td>0.623966</td>\n",
       "      <td>0.344200</td>\n",
       "      <td>0.657303</td>\n",
       "      <td>0.999949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149999</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.005469</td>\n",
       "      <td>0.264093</td>\n",
       "      <td>0.278099</td>\n",
       "      <td>0.790717</td>\n",
       "      <td>0.507065</td>\n",
       "      <td>0.526442</td>\n",
       "      <td>0.415497</td>\n",
       "      <td>0.190937</td>\n",
       "      <td>0.566858</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612245</td>\n",
       "      <td>0.431227</td>\n",
       "      <td>0.428153</td>\n",
       "      <td>0.300445</td>\n",
       "      <td>0.562332</td>\n",
       "      <td>0.380863</td>\n",
       "      <td>0.505085</td>\n",
       "      <td>0.353031</td>\n",
       "      <td>0.661554</td>\n",
       "      <td>0.999909</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150000 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Ocorrencia       PP1       PP2       PP3       PP4       PP5  \\\n",
       "0         1.000000  0.060108  0.269816  0.194538  0.759211  0.525677   \n",
       "1         0.999989  0.124743  0.280167  0.211427  0.797845  0.535842   \n",
       "2         0.999978  0.437922  0.298739  0.201101  0.729084  0.551182   \n",
       "3         0.999978  0.071959  0.257045  0.212612  0.706865  0.510530   \n",
       "4         0.999966  0.659764  0.288978  0.238128  0.764649  0.534751   \n",
       "5         0.999966  0.060939  0.257608  0.202888  0.694450  0.520002   \n",
       "6         0.999966  0.063501  0.254961  0.214156  0.709450  0.509648   \n",
       "7         0.999966  0.628501  0.314227  0.196994  0.763221  0.470152   \n",
       "8         0.999966  0.058028  0.263326  0.209173  0.811448  0.524605   \n",
       "9         0.999955  0.503509  0.263252  0.182764  0.818657  0.529358   \n",
       "10        0.999944  0.507369  0.249225  0.177989  0.775231  0.509743   \n",
       "11        0.999944  0.068074  0.259608  0.186721  0.694350  0.529656   \n",
       "12        0.999933  0.581976  0.270174  0.166837  0.730940  0.488154   \n",
       "13        0.999921  0.045959  0.265375  0.216054  0.800336  0.518120   \n",
       "14        0.999910  0.055022  0.269408  0.247157  0.826264  0.490317   \n",
       "15        0.999910  0.053504  0.268520  0.197594  0.836429  0.540564   \n",
       "16        0.999910  0.061692  0.257787  0.223630  0.713750  0.509586   \n",
       "17        0.999899  0.491944  0.247020  0.210613  0.662371  0.485535   \n",
       "18        0.999899  0.068406  0.259734  0.206104  0.714286  0.521034   \n",
       "19        0.999899  0.091646  0.263529  0.212942  0.687926  0.514817   \n",
       "20        0.999888  0.603368  0.245626  0.193507  0.778744  0.538936   \n",
       "21        0.999876  0.547288  0.263441  0.146722  0.742337  0.516630   \n",
       "22        0.999876  0.070490  0.263707  0.243113  0.813920  0.501833   \n",
       "23        0.999865  0.064571  0.272311  0.196872  0.774924  0.537216   \n",
       "24        0.999854  0.622654  0.302666  0.166313  0.770253  0.514497   \n",
       "25        0.999843  0.468042  0.246626  0.179216  0.748115  0.512977   \n",
       "26        0.999843  0.068820  0.259389  0.225439  0.740207  0.515899   \n",
       "27        0.999843  0.078209  0.261978  0.198220  0.683821  0.522798   \n",
       "28        0.999831  0.056515  0.254350  0.221146  0.734918  0.507038   \n",
       "29        0.999831  0.067138  0.269948  0.197098  0.783340  0.529933   \n",
       "...            ...       ...       ...       ...       ...       ...   \n",
       "149970    0.000157  0.532614  0.233550  0.267199  0.781760  0.505738   \n",
       "149971    0.000157  0.046490  0.264995  0.270647  0.702956  0.491346   \n",
       "149972    0.000157  0.481101  0.265155  0.245330  0.747723  0.507301   \n",
       "149973    0.000146  0.508196  0.247321  0.217955  0.798762  0.500127   \n",
       "149974    0.000146  0.470533  0.254848  0.180369  0.783546  0.503923   \n",
       "149975    0.000135  0.409908  0.253989  0.195949  0.788477  0.509173   \n",
       "149976    0.000124  0.182477  0.247111  0.232192  0.782715  0.493247   \n",
       "149977    0.000124  0.019006  0.261519  0.243387  0.722387  0.515834   \n",
       "149978    0.000112  0.042613  0.271372  0.255537  0.730603  0.487255   \n",
       "149979    0.000112  0.571725  0.276648  0.218708  0.835705  0.489958   \n",
       "149980    0.000112  0.200347  0.246626  0.225561  0.776334  0.497161   \n",
       "149981    0.000101  0.021856  0.263580  0.222621  0.725007  0.524848   \n",
       "149982    0.000090  0.567706  0.229124  0.265600  0.770859  0.519726   \n",
       "149983    0.000079  0.028427  0.259189  0.268290  0.694166  0.499947   \n",
       "149984    0.000067  0.186562  0.254587  0.198728  0.723938  0.518759   \n",
       "149985    0.000067  0.202279  0.246037  0.242916  0.746867  0.509363   \n",
       "149986    0.000067  0.065040  0.279523  0.221239  0.700656  0.533802   \n",
       "149987    0.000067  0.171513  0.245399  0.241619  0.747931  0.508853   \n",
       "149988    0.000056  0.533147  0.241624  0.228711  0.827618  0.493825   \n",
       "149989    0.000056  0.026445  0.262856  0.241106  0.724946  0.516910   \n",
       "149990    0.000045  0.462767  0.245008  0.206387  0.721055  0.496705   \n",
       "149991    0.000045  0.491811  0.252618  0.212195  0.690086  0.520855   \n",
       "149992    0.000045  0.019255  0.259797  0.266591  0.741629  0.502813   \n",
       "149993    0.000034  0.018064  0.263096  0.278815  0.778208  0.474551   \n",
       "149994    0.000034  0.017594  0.276054  0.258699  0.828557  0.495146   \n",
       "149995    0.000011  0.497450  0.241053  0.247879  0.778489  0.507282   \n",
       "149996    0.000000  0.390792  0.235442  0.204946  0.569748  0.504264   \n",
       "149997    0.000000  0.568302  0.247759  0.227228  0.756020  0.507406   \n",
       "149998    0.000000  0.521388  0.247262  0.179510  0.764294  0.516755   \n",
       "149999    0.000000  0.005469  0.264093  0.278099  0.790717  0.507065   \n",
       "\n",
       "             PP6       PP7       PP8       PP9    ...         PP20      PP21  \\\n",
       "0       0.454007  0.439219  0.179817  0.536894    ...     0.615605  0.444771   \n",
       "1       0.498879  0.423118  0.185079  0.480048    ...     0.598063  0.432270   \n",
       "2       0.480305  0.424487  0.184995  0.529176    ...     0.583944  0.428361   \n",
       "3       0.485870  0.420289  0.182855  0.545253    ...     0.612383  0.436144   \n",
       "4       0.457382  0.417722  0.170328  0.525486    ...     0.621350  0.442086   \n",
       "5       0.498166  0.423302  0.184329  0.509488    ...     0.613390  0.444842   \n",
       "6       0.492359  0.419326  0.184466  0.547298    ...     0.612069  0.437132   \n",
       "7       0.511660  0.437191  0.179676  0.576492    ...     0.589196  0.434704   \n",
       "8       0.506637  0.424924  0.184904  0.484689    ...     0.611292  0.443771   \n",
       "9       0.504508  0.408831  0.185765  0.594448    ...     0.610703  0.438246   \n",
       "10      0.495201  0.412585  0.184154  0.537584    ...     0.612686  0.437830   \n",
       "11      0.495904  0.429824  0.181994  0.512852    ...     0.613391  0.437905   \n",
       "12      0.472377  0.439546  0.177985  0.507370    ...     0.625896  0.440003   \n",
       "13      0.483458  0.431821  0.183742  0.563240    ...     0.607720  0.438863   \n",
       "14      0.418498  0.432895  0.176874  0.569615    ...     0.604352  0.436274   \n",
       "15      0.522830  0.431134  0.187447  0.522275    ...     0.616231  0.443603   \n",
       "16      0.492926  0.419698  0.184235  0.534748    ...     0.611983  0.438476   \n",
       "17      0.384332  0.430031  0.168089  0.572563    ...     0.607939  0.435439   \n",
       "18      0.501808  0.421213  0.184896  0.528196    ...     0.610687  0.441122   \n",
       "19      0.478196  0.420261  0.181587  0.529101    ...     0.610140  0.437772   \n",
       "20      0.508206  0.430635  0.169207  0.528025    ...     0.609587  0.436557   \n",
       "21      0.478693  0.429831  0.177433  0.497740    ...     0.615299  0.440121   \n",
       "22      0.488295  0.413711  0.186118  0.515571    ...     0.605819  0.443106   \n",
       "23      0.492495  0.436480  0.182498  0.541370    ...     0.607089  0.432516   \n",
       "24      0.513362  0.423665  0.181601  0.556268    ...     0.580306  0.429427   \n",
       "25      0.506617  0.410903  0.184700  0.552606    ...     0.608547  0.441297   \n",
       "26      0.524262  0.411517  0.190307  0.546224    ...     0.606831  0.438082   \n",
       "27      0.487356  0.424311  0.183387  0.506036    ...     0.610572  0.438074   \n",
       "28      0.494405  0.419691  0.185488  0.546602    ...     0.609531  0.443414   \n",
       "29      0.479951  0.436058  0.181232  0.559892    ...     0.607504  0.430085   \n",
       "...          ...       ...       ...       ...    ...          ...       ...   \n",
       "149970  0.503238  0.421217  0.171538  0.543447    ...     0.609687  0.443950   \n",
       "149971  0.467347  0.411664  0.182871  0.549368    ...     0.610117  0.432241   \n",
       "149972  0.490690  0.411432  0.183605  0.521229    ...     0.612440  0.436974   \n",
       "149973  0.510187  0.407694  0.185739  0.548191    ...     0.614358  0.439940   \n",
       "149974  0.467158  0.417149  0.181294  0.513165    ...     0.610052  0.434837   \n",
       "149975  0.492489  0.414502  0.183078  0.536575    ...     0.610661  0.439295   \n",
       "149976  0.497517  0.408425  0.184280  0.545814    ...     0.609134  0.443769   \n",
       "149977  0.513095  0.419868  0.187521  0.495593    ...     0.614397  0.437045   \n",
       "149978  0.395535  0.433196  0.173116  0.506258    ...     0.606196  0.437104   \n",
       "149979  0.506151  0.419653  0.185217  0.571860    ...     0.606836  0.439324   \n",
       "149980  0.504085  0.408024  0.184750  0.549782    ...     0.609117  0.442888   \n",
       "149981  0.506250  0.427495  0.186490  0.498180    ...     0.611260  0.434648   \n",
       "149982  0.515122  0.425296  0.168247  0.545042    ...     0.610050  0.431271   \n",
       "149983  0.513015  0.407354  0.190131  0.543477    ...     0.609372  0.435329   \n",
       "149984  0.460384  0.405285  0.191462  0.514548    ...     0.604728  0.436449   \n",
       "149985  0.522950  0.407422  0.185101  0.550593    ...     0.609743  0.432266   \n",
       "149986  0.485056  0.426519  0.183214  0.487101    ...     0.601667  0.437088   \n",
       "149987  0.526252  0.411864  0.183825  0.541090    ...     0.613549  0.433180   \n",
       "149988  0.494891  0.406833  0.183603  0.541563    ...     0.615864  0.439238   \n",
       "149989  0.514813  0.418072  0.188466  0.507721    ...     0.610454  0.437349   \n",
       "149990  0.503385  0.396524  0.188572  0.575909    ...     0.607776  0.436937   \n",
       "149991  0.472870  0.408565  0.180919  0.535867    ...     0.611110  0.444796   \n",
       "149992  0.513255  0.411977  0.188803  0.540003    ...     0.611673  0.435822   \n",
       "149993  0.415605  0.425636  0.176473  0.522033    ...     0.611982  0.442510   \n",
       "149994  0.403561  0.443447  0.173894  0.530968    ...     0.607534  0.431134   \n",
       "149995  0.500286  0.419056  0.174705  0.543771    ...     0.612283  0.431614   \n",
       "149996  0.480996  0.387067  0.199926  0.556866    ...     0.594849  0.439691   \n",
       "149997  0.495916  0.402733  0.190475  0.526601    ...     0.614769  0.444031   \n",
       "149998  0.481496  0.421304  0.175882  0.525186    ...     0.610705  0.444292   \n",
       "149999  0.526442  0.415497  0.190937  0.566858    ...     0.612245  0.431227   \n",
       "\n",
       "            PP22      PP23      PP24      PP25      PP26      PP27      PP28  \\\n",
       "0       0.503320  0.299819  0.908043  0.405191  0.437251  0.348026  0.658603   \n",
       "1       0.469694  0.305838  0.669805  0.402392  0.540172  0.351041  0.656548   \n",
       "2       0.471864  0.308728  0.599245  0.418174  0.586579  0.351533  0.654027   \n",
       "3       0.470058  0.298905  0.737281  0.394715  0.602328  0.349004  0.658670   \n",
       "4       0.527476  0.318317  0.946516  0.436698  0.404923  0.346270  0.688703   \n",
       "5       0.537481  0.296726  0.721367  0.399994  0.656492  0.348901  0.658189   \n",
       "6       0.475351  0.299724  0.738411  0.386322  0.603711  0.349224  0.658782   \n",
       "7       0.512702  0.275804  0.925919  0.425253  0.494546  0.353528  0.652259   \n",
       "8       0.523454  0.297067  0.684544  0.401863  0.601175  0.349428  0.658710   \n",
       "9       0.486868  0.293551  0.594133  0.424509  0.386004  0.351862  0.656021   \n",
       "10      0.483856  0.303515  0.678457  0.407128  0.490479  0.357300  0.656700   \n",
       "11      0.482166  0.297311  0.604921  0.406035  0.625691  0.348374  0.658111   \n",
       "12      0.466206  0.307770  0.837685  0.464918  0.485192  0.344128  0.659633   \n",
       "13      0.498374  0.300060  0.927711  0.396627  0.600367  0.349539  0.658965   \n",
       "14      0.489494  0.300122  0.512642  0.386626  0.587311  0.350241  0.658232   \n",
       "15      0.505097  0.296043  0.565238  0.412659  0.569274  0.348296  0.657760   \n",
       "16      0.496585  0.301951  0.838281  0.380579  0.599394  0.350318  0.658867   \n",
       "17      0.469519  0.300365  0.510056  0.432115  0.495633  0.351272  0.656911   \n",
       "18      0.509858  0.298266  0.628529  0.399882  0.510058  0.351072  0.658633   \n",
       "19      0.491034  0.300148  0.740123  0.396474  0.606181  0.349557  0.658190   \n",
       "20      0.478335  0.301043  0.606673  0.397660  0.370018  0.345905  0.661565   \n",
       "21      0.490750  0.302470  0.647573  0.412102  0.500056  0.350303  0.657476   \n",
       "22      0.537544  0.301911  0.980834  0.384053  0.712976  0.349469  0.658396   \n",
       "23      0.450391  0.301169  0.684242  0.400202  0.561055  0.349462  0.658185   \n",
       "24      0.476694  0.275411  0.589198  0.385847  0.609710  0.347876  0.652595   \n",
       "25      0.509048  0.296762  0.623086  0.439052  0.539124  0.341961  0.655334   \n",
       "26      0.495486  0.300805  0.602424  0.392221  0.371565  0.353787  0.658535   \n",
       "27      0.479237  0.300134  0.670161  0.390377  0.596555  0.348346  0.657806   \n",
       "28      0.530226  0.298384  0.837443  0.401339  0.527562  0.350686  0.658582   \n",
       "29      0.426398  0.299435  0.728297  0.409457  0.555204  0.348438  0.658332   \n",
       "...          ...       ...       ...       ...       ...       ...       ...   \n",
       "149970  0.539304  0.295740  0.759702  0.427636  0.520482  0.348044  0.659008   \n",
       "149971  0.449527  0.299557  0.977777  0.417670  0.617311  0.350228  0.660544   \n",
       "149972  0.469025  0.294617  0.632493  0.544335  0.610601  0.345107  0.664333   \n",
       "149973  0.503672  0.297323  0.743479  0.472405  0.569249  0.366154  0.661006   \n",
       "149974  0.435652  0.300087  0.858866  0.455757  0.445736  0.342710  0.659314   \n",
       "149975  0.500524  0.292375  0.561481  0.504029  0.551778  0.344917  0.651831   \n",
       "149976  0.528883  0.297417  0.691677  0.443563  0.530746  0.343813  0.656884   \n",
       "149977  0.464115  0.297523  0.698902  0.408582  0.575995  0.350133  0.660582   \n",
       "149978  0.497376  0.295048  0.576284  0.443841  0.694414  0.348511  0.658689   \n",
       "149979  0.494412  0.306040  0.933722  0.360417  0.565301  0.343219  0.659189   \n",
       "149980  0.520362  0.296112  0.584495  0.448522  0.537186  0.343611  0.656676   \n",
       "149981  0.446650  0.296231  0.664746  0.433839  0.446474  0.350519  0.660111   \n",
       "149982  0.432195  0.295865  0.565957  0.450771  0.591798  0.340308  0.651638   \n",
       "149983  0.474033  0.299532  0.584335  0.397104  0.642554  0.351759  0.659867   \n",
       "149984  0.435638  0.300399  0.673756  0.469924  0.481569  0.355139  0.667192   \n",
       "149985  0.440902  0.296688  0.688173  0.433036  0.580197  0.352513  0.659998   \n",
       "149986  0.501360  0.297700  0.679562  0.458239  0.542116  0.351499  0.658380   \n",
       "149987  0.447016  0.298735  0.694610  0.437655  0.578895  0.352183  0.660237   \n",
       "149988  0.508968  0.301390  0.926773  0.424499  0.690206  0.369836  0.655819   \n",
       "149989  0.478786  0.296534  0.687449  0.423486  0.612608  0.350272  0.659835   \n",
       "149990  0.475566  0.303448  0.496558  0.347448  0.611472  0.352421  0.660536   \n",
       "149991  0.535423  0.289715  0.593682  0.454752  0.730033  0.349653  0.659670   \n",
       "149992  0.466025  0.298341  0.542823  0.404320  0.421093  0.354049  0.661279   \n",
       "149993  0.521803  0.292841  0.566563  0.431265  0.501955  0.351654  0.660878   \n",
       "149994  0.431869  0.296627  0.557982  0.426925  0.546044  0.349030  0.660171   \n",
       "149995  0.435018  0.298214  0.651996  0.461655  0.591658  0.339955  0.651693   \n",
       "149996  0.425836  0.300856  0.676339  0.487193  0.506982  0.341864  0.675988   \n",
       "149997  0.497298  0.298596  0.761663  0.455216  0.504276  0.371058  0.663423   \n",
       "149998  0.529266  0.301788  0.776543  0.396480  0.623966  0.344200  0.657303   \n",
       "149999  0.428153  0.300445  0.562332  0.380863  0.505085  0.353031  0.661554   \n",
       "\n",
       "          Sacado  \n",
       "0       0.998556  \n",
       "1       0.979294  \n",
       "2       0.959301  \n",
       "3       0.998409  \n",
       "4       0.973436  \n",
       "5       0.999885  \n",
       "6       0.999924  \n",
       "7       0.980414  \n",
       "8       0.999942  \n",
       "9       0.989651  \n",
       "10      0.999833  \n",
       "11      0.999940  \n",
       "12      0.999080  \n",
       "13      0.999686  \n",
       "14      0.995587  \n",
       "15      0.998972  \n",
       "16      0.998008  \n",
       "17      0.997636  \n",
       "18      0.997664  \n",
       "19      0.992420  \n",
       "20      0.997456  \n",
       "21      0.998812  \n",
       "22      0.992433  \n",
       "23      0.995113  \n",
       "24      0.974552  \n",
       "25      0.999416  \n",
       "26      0.993946  \n",
       "27      0.995621  \n",
       "28      0.999955  \n",
       "29      0.996439  \n",
       "...          ...  \n",
       "149970  0.999541  \n",
       "149971  0.989366  \n",
       "149972  0.990151  \n",
       "149973  0.999543  \n",
       "149974  0.999415  \n",
       "149975  0.998123  \n",
       "149976  0.999695  \n",
       "149977  0.999302  \n",
       "149978  0.989566  \n",
       "149979  0.988276  \n",
       "149980  0.999695  \n",
       "149981  0.998474  \n",
       "149982  0.999568  \n",
       "149983  0.992674  \n",
       "149984  0.989825  \n",
       "149985  0.995663  \n",
       "149986  0.980729  \n",
       "149987  0.998917  \n",
       "149988  0.999401  \n",
       "149989  0.995236  \n",
       "149990  0.996347  \n",
       "149991  0.988799  \n",
       "149992  0.997692  \n",
       "149993  0.999398  \n",
       "149994  0.996959  \n",
       "149995  0.999950  \n",
       "149996  0.992286  \n",
       "149997  0.994495  \n",
       "149998  0.999949  \n",
       "149999  0.999909  \n",
       "\n",
       "[150000 rows x 30 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_scaled = df[features].copy()\n",
    "scaler = MinMaxScaler()\n",
    "\n",
    "df_scaled = pd.DataFrame(scaler.fit_transform(df_scaled), columns=features)\n",
    "\n",
    "df_scaled"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Variances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Variance (Scaled)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PP28</th>\n",
       "      <td>0.000080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP23</th>\n",
       "      <td>0.000095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP27</th>\n",
       "      <td>0.000131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP20</th>\n",
       "      <td>0.000139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP21</th>\n",
       "      <td>0.000142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sacado</th>\n",
       "      <td>0.000158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP8</th>\n",
       "      <td>0.000180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP7</th>\n",
       "      <td>0.000253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP2</th>\n",
       "      <td>0.000361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP5</th>\n",
       "      <td>0.000436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP10</th>\n",
       "      <td>0.000495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP17</th>\n",
       "      <td>0.000712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP25</th>\n",
       "      <td>0.000807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP6</th>\n",
       "      <td>0.000919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP14</th>\n",
       "      <td>0.000992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP3</th>\n",
       "      <td>0.001066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP22</th>\n",
       "      <td>0.001090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP12</th>\n",
       "      <td>0.001371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP9</th>\n",
       "      <td>0.001452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP16</th>\n",
       "      <td>0.001935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP18</th>\n",
       "      <td>0.003342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP4</th>\n",
       "      <td>0.003838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP19</th>\n",
       "      <td>0.004088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP11</th>\n",
       "      <td>0.004947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP26</th>\n",
       "      <td>0.006898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP13</th>\n",
       "      <td>0.009267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP15</th>\n",
       "      <td>0.009901</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP24</th>\n",
       "      <td>0.010583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP1</th>\n",
       "      <td>0.055678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ocorrencia</th>\n",
       "      <td>0.097080</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Variance (Scaled)\n",
       "PP28                 0.000080\n",
       "PP23                 0.000095\n",
       "PP27                 0.000131\n",
       "PP20                 0.000139\n",
       "PP21                 0.000142\n",
       "Sacado               0.000158\n",
       "PP8                  0.000180\n",
       "PP7                  0.000253\n",
       "PP2                  0.000361\n",
       "PP5                  0.000436\n",
       "PP10                 0.000495\n",
       "PP17                 0.000712\n",
       "PP25                 0.000807\n",
       "PP6                  0.000919\n",
       "PP14                 0.000992\n",
       "PP3                  0.001066\n",
       "PP22                 0.001090\n",
       "PP12                 0.001371\n",
       "PP9                  0.001452\n",
       "PP16                 0.001935\n",
       "PP18                 0.003342\n",
       "PP4                  0.003838\n",
       "PP19                 0.004088\n",
       "PP11                 0.004947\n",
       "PP26                 0.006898\n",
       "PP13                 0.009267\n",
       "PP15                 0.009901\n",
       "PP24                 0.010583\n",
       "PP1                  0.055678\n",
       "Ocorrencia           0.097080"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selector = VarianceThreshold()\n",
    "\n",
    "selector.fit(df_scaled)\n",
    "\n",
    "df_var = pd.DataFrame({'Variance (Scaled)': selector.variances_}, index=features)\n",
    "\n",
    "df_var.sort_values(by='Variance (Scaled)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PP23, PP28 and PP27 have the least variance and (looking at the correlations heatmap) no correlation to the Target - will drop those features. PP20 has correlation with Sacado, so I will leave it."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Remove Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(['PP23', 'PP28', 'PP27'], axis=1, inplace=True)\n",
    "\n",
    "features = [col for col in df if col != 'Fraude']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PCA <a class=\"anchor\" id=\"7\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For PCA, we need to first scale the data with Standardization (mean=0 and variance=1)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ocorrencia</th>\n",
       "      <th>PP1</th>\n",
       "      <th>PP2</th>\n",
       "      <th>PP3</th>\n",
       "      <th>PP4</th>\n",
       "      <th>PP5</th>\n",
       "      <th>PP6</th>\n",
       "      <th>PP7</th>\n",
       "      <th>PP8</th>\n",
       "      <th>PP9</th>\n",
       "      <th>...</th>\n",
       "      <th>PP18</th>\n",
       "      <th>PP19</th>\n",
       "      <th>PP20</th>\n",
       "      <th>PP21</th>\n",
       "      <th>PP22</th>\n",
       "      <th>PP24</th>\n",
       "      <th>PP25</th>\n",
       "      <th>PP26</th>\n",
       "      <th>Sacado</th>\n",
       "      <th>Fraude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.452554</td>\n",
       "      <td>-0.985473</td>\n",
       "      <td>0.607243</td>\n",
       "      <td>-0.578142</td>\n",
       "      <td>0.206595</td>\n",
       "      <td>0.604854</td>\n",
       "      <td>-1.260358</td>\n",
       "      <td>1.192228</td>\n",
       "      <td>-0.393617</td>\n",
       "      <td>-0.033998</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.785697</td>\n",
       "      <td>0.216948</td>\n",
       "      <td>0.503995</td>\n",
       "      <td>0.510952</td>\n",
       "      <td>0.365689</td>\n",
       "      <td>2.141398</td>\n",
       "      <td>-0.525371</td>\n",
       "      <td>-1.389421</td>\n",
       "      <td>0.243518</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.452518</td>\n",
       "      <td>-0.711549</td>\n",
       "      <td>1.151666</td>\n",
       "      <td>-0.060850</td>\n",
       "      <td>0.830176</td>\n",
       "      <td>1.091942</td>\n",
       "      <td>0.219741</td>\n",
       "      <td>0.180462</td>\n",
       "      <td>-0.001535</td>\n",
       "      <td>-1.525840</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.085263</td>\n",
       "      <td>-1.182226</td>\n",
       "      <td>-0.984745</td>\n",
       "      <td>-0.537851</td>\n",
       "      <td>-0.652846</td>\n",
       "      <td>-0.174406</td>\n",
       "      <td>-0.623881</td>\n",
       "      <td>-0.150271</td>\n",
       "      <td>-1.287488</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.452481</td>\n",
       "      <td>0.615691</td>\n",
       "      <td>2.128527</td>\n",
       "      <td>-0.377129</td>\n",
       "      <td>-0.279669</td>\n",
       "      <td>1.826966</td>\n",
       "      <td>-0.392920</td>\n",
       "      <td>0.266475</td>\n",
       "      <td>-0.007794</td>\n",
       "      <td>-0.236547</td>\n",
       "      <td>...</td>\n",
       "      <td>2.331057</td>\n",
       "      <td>-0.279733</td>\n",
       "      <td>-2.183033</td>\n",
       "      <td>-0.865822</td>\n",
       "      <td>-0.587094</td>\n",
       "      <td>-0.860282</td>\n",
       "      <td>-0.068399</td>\n",
       "      <td>0.408465</td>\n",
       "      <td>-2.876641</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.452481</td>\n",
       "      <td>-0.935247</td>\n",
       "      <td>-0.064476</td>\n",
       "      <td>-0.024557</td>\n",
       "      <td>-0.638310</td>\n",
       "      <td>-0.120940</td>\n",
       "      <td>-0.209379</td>\n",
       "      <td>0.002692</td>\n",
       "      <td>-0.167246</td>\n",
       "      <td>0.185366</td>\n",
       "      <td>...</td>\n",
       "      <td>0.600185</td>\n",
       "      <td>0.981217</td>\n",
       "      <td>0.230558</td>\n",
       "      <td>-0.212839</td>\n",
       "      <td>-0.641821</td>\n",
       "      <td>0.481502</td>\n",
       "      <td>-0.894097</td>\n",
       "      <td>0.598090</td>\n",
       "      <td>0.231791</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>1.555849</td>\n",
       "      <td>1.615105</td>\n",
       "      <td>0.756921</td>\n",
       "      <td>0.294379</td>\n",
       "      <td>1.039644</td>\n",
       "      <td>-1.149025</td>\n",
       "      <td>-0.158615</td>\n",
       "      <td>-1.100691</td>\n",
       "      <td>-0.333398</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.319784</td>\n",
       "      <td>-0.443985</td>\n",
       "      <td>0.991586</td>\n",
       "      <td>0.285680</td>\n",
       "      <td>1.097367</td>\n",
       "      <td>2.515379</td>\n",
       "      <td>0.583625</td>\n",
       "      <td>-1.778647</td>\n",
       "      <td>-1.753154</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>-0.981952</td>\n",
       "      <td>-0.034877</td>\n",
       "      <td>-0.322386</td>\n",
       "      <td>-0.838697</td>\n",
       "      <td>0.332925</td>\n",
       "      <td>0.196228</td>\n",
       "      <td>0.192049</td>\n",
       "      <td>-0.057406</td>\n",
       "      <td>-0.753223</td>\n",
       "      <td>...</td>\n",
       "      <td>0.265489</td>\n",
       "      <td>-0.316328</td>\n",
       "      <td>0.316037</td>\n",
       "      <td>0.516861</td>\n",
       "      <td>1.400395</td>\n",
       "      <td>0.326806</td>\n",
       "      <td>-0.708302</td>\n",
       "      <td>1.250213</td>\n",
       "      <td>0.349097</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>-0.971092</td>\n",
       "      <td>-0.174094</td>\n",
       "      <td>0.022727</td>\n",
       "      <td>-0.596579</td>\n",
       "      <td>-0.163207</td>\n",
       "      <td>0.004658</td>\n",
       "      <td>-0.057781</td>\n",
       "      <td>-0.047207</td>\n",
       "      <td>0.239018</td>\n",
       "      <td>...</td>\n",
       "      <td>0.013302</td>\n",
       "      <td>0.376667</td>\n",
       "      <td>0.203881</td>\n",
       "      <td>-0.129952</td>\n",
       "      <td>-0.481487</td>\n",
       "      <td>0.492480</td>\n",
       "      <td>-1.189520</td>\n",
       "      <td>0.614738</td>\n",
       "      <td>0.352211</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>1.423358</td>\n",
       "      <td>2.943124</td>\n",
       "      <td>-0.502896</td>\n",
       "      <td>0.271327</td>\n",
       "      <td>-2.055703</td>\n",
       "      <td>0.641306</td>\n",
       "      <td>1.064781</td>\n",
       "      <td>-0.404160</td>\n",
       "      <td>1.005183</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.141295</td>\n",
       "      <td>1.232680</td>\n",
       "      <td>-1.737285</td>\n",
       "      <td>-0.333678</td>\n",
       "      <td>0.649857</td>\n",
       "      <td>2.315159</td>\n",
       "      <td>0.180772</td>\n",
       "      <td>-0.699595</td>\n",
       "      <td>-1.198487</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.452445</td>\n",
       "      <td>-0.994288</td>\n",
       "      <td>0.265920</td>\n",
       "      <td>-0.129913</td>\n",
       "      <td>1.049748</td>\n",
       "      <td>0.553487</td>\n",
       "      <td>0.475634</td>\n",
       "      <td>0.293952</td>\n",
       "      <td>-0.014615</td>\n",
       "      <td>-1.404025</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.079015</td>\n",
       "      <td>-1.945638</td>\n",
       "      <td>0.137952</td>\n",
       "      <td>0.426987</td>\n",
       "      <td>0.975534</td>\n",
       "      <td>-0.031139</td>\n",
       "      <td>-0.642520</td>\n",
       "      <td>0.584208</td>\n",
       "      <td>0.353666</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.452409</td>\n",
       "      <td>0.893649</td>\n",
       "      <td>0.262019</td>\n",
       "      <td>-0.938725</td>\n",
       "      <td>1.166103</td>\n",
       "      <td>0.781218</td>\n",
       "      <td>0.405424</td>\n",
       "      <td>-0.717286</td>\n",
       "      <td>0.049532</td>\n",
       "      <td>1.476413</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.134925</td>\n",
       "      <td>1.511254</td>\n",
       "      <td>0.087955</td>\n",
       "      <td>-0.036519</td>\n",
       "      <td>-0.132637</td>\n",
       "      <td>-0.909979</td>\n",
       "      <td>0.154585</td>\n",
       "      <td>-2.006428</td>\n",
       "      <td>-0.464282</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.452373</td>\n",
       "      <td>0.910005</td>\n",
       "      <td>-0.475790</td>\n",
       "      <td>-1.084986</td>\n",
       "      <td>0.465173</td>\n",
       "      <td>-0.158645</td>\n",
       "      <td>0.098420</td>\n",
       "      <td>-0.481386</td>\n",
       "      <td>-0.070508</td>\n",
       "      <td>-0.015903</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.162083</td>\n",
       "      <td>0.000508</td>\n",
       "      <td>0.256279</td>\n",
       "      <td>-0.071417</td>\n",
       "      <td>-0.223865</td>\n",
       "      <td>-0.090307</td>\n",
       "      <td>-0.457207</td>\n",
       "      <td>-0.748568</td>\n",
       "      <td>0.344973</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.452373</td>\n",
       "      <td>-0.951713</td>\n",
       "      <td>0.070342</td>\n",
       "      <td>-0.817547</td>\n",
       "      <td>-0.840311</td>\n",
       "      <td>0.795529</td>\n",
       "      <td>0.121590</td>\n",
       "      <td>0.601885</td>\n",
       "      <td>-0.231436</td>\n",
       "      <td>-0.664958</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.838740</td>\n",
       "      <td>0.160756</td>\n",
       "      <td>0.316081</td>\n",
       "      <td>-0.065126</td>\n",
       "      <td>-0.275068</td>\n",
       "      <td>-0.805109</td>\n",
       "      <td>-0.495667</td>\n",
       "      <td>0.879377</td>\n",
       "      <td>0.353505</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.452337</td>\n",
       "      <td>1.226187</td>\n",
       "      <td>0.626067</td>\n",
       "      <td>-1.426541</td>\n",
       "      <td>-0.249718</td>\n",
       "      <td>-1.193089</td>\n",
       "      <td>-0.654438</td>\n",
       "      <td>1.212763</td>\n",
       "      <td>-0.530135</td>\n",
       "      <td>-0.808822</td>\n",
       "      <td>...</td>\n",
       "      <td>2.410552</td>\n",
       "      <td>1.512248</td>\n",
       "      <td>1.377398</td>\n",
       "      <td>0.110948</td>\n",
       "      <td>-0.758486</td>\n",
       "      <td>1.457480</td>\n",
       "      <td>1.576930</td>\n",
       "      <td>-0.812224</td>\n",
       "      <td>0.285167</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.452301</td>\n",
       "      <td>-1.045434</td>\n",
       "      <td>0.373695</td>\n",
       "      <td>0.080851</td>\n",
       "      <td>0.870383</td>\n",
       "      <td>0.242749</td>\n",
       "      <td>-0.288918</td>\n",
       "      <td>0.727344</td>\n",
       "      <td>-0.101179</td>\n",
       "      <td>0.657392</td>\n",
       "      <td>...</td>\n",
       "      <td>0.312318</td>\n",
       "      <td>-1.743823</td>\n",
       "      <td>-0.165211</td>\n",
       "      <td>0.015303</td>\n",
       "      <td>0.215884</td>\n",
       "      <td>2.332584</td>\n",
       "      <td>-0.826810</td>\n",
       "      <td>0.574471</td>\n",
       "      <td>0.333327</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.452265</td>\n",
       "      <td>-1.007026</td>\n",
       "      <td>0.585796</td>\n",
       "      <td>1.033456</td>\n",
       "      <td>1.288880</td>\n",
       "      <td>-1.089464</td>\n",
       "      <td>-2.431620</td>\n",
       "      <td>0.794830</td>\n",
       "      <td>-0.612916</td>\n",
       "      <td>0.824701</td>\n",
       "      <td>...</td>\n",
       "      <td>1.105854</td>\n",
       "      <td>-1.066890</td>\n",
       "      <td>-0.451001</td>\n",
       "      <td>-0.201954</td>\n",
       "      <td>-0.053102</td>\n",
       "      <td>-1.702112</td>\n",
       "      <td>-1.178848</td>\n",
       "      <td>0.417280</td>\n",
       "      <td>0.007530</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1.452265</td>\n",
       "      <td>-1.013460</td>\n",
       "      <td>0.539064</td>\n",
       "      <td>-0.484548</td>\n",
       "      <td>1.452957</td>\n",
       "      <td>1.318202</td>\n",
       "      <td>1.009762</td>\n",
       "      <td>0.684201</td>\n",
       "      <td>0.174900</td>\n",
       "      <td>-0.417668</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.629926</td>\n",
       "      <td>0.420063</td>\n",
       "      <td>0.557118</td>\n",
       "      <td>0.412909</td>\n",
       "      <td>0.419496</td>\n",
       "      <td>-1.190853</td>\n",
       "      <td>-0.262500</td>\n",
       "      <td>0.200119</td>\n",
       "      <td>0.276595</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1.452265</td>\n",
       "      <td>-0.978759</td>\n",
       "      <td>-0.025438</td>\n",
       "      <td>0.312878</td>\n",
       "      <td>-0.527179</td>\n",
       "      <td>-0.166151</td>\n",
       "      <td>0.023364</td>\n",
       "      <td>-0.034418</td>\n",
       "      <td>-0.064410</td>\n",
       "      <td>-0.090317</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.171218</td>\n",
       "      <td>-0.449572</td>\n",
       "      <td>0.196602</td>\n",
       "      <td>-0.017233</td>\n",
       "      <td>0.161675</td>\n",
       "      <td>1.463277</td>\n",
       "      <td>-1.391687</td>\n",
       "      <td>0.562756</td>\n",
       "      <td>0.199927</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1.452229</td>\n",
       "      <td>0.844634</td>\n",
       "      <td>-0.591753</td>\n",
       "      <td>-0.085803</td>\n",
       "      <td>-1.356474</td>\n",
       "      <td>-1.318614</td>\n",
       "      <td>-3.558593</td>\n",
       "      <td>0.614877</td>\n",
       "      <td>-1.267491</td>\n",
       "      <td>0.902074</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.892266</td>\n",
       "      <td>-1.582167</td>\n",
       "      <td>-0.146567</td>\n",
       "      <td>-0.271966</td>\n",
       "      <td>-0.658121</td>\n",
       "      <td>-1.727251</td>\n",
       "      <td>0.422321</td>\n",
       "      <td>-0.686513</td>\n",
       "      <td>0.170368</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1.452229</td>\n",
       "      <td>-0.950305</td>\n",
       "      <td>0.076973</td>\n",
       "      <td>-0.223885</td>\n",
       "      <td>-0.518524</td>\n",
       "      <td>0.382386</td>\n",
       "      <td>0.316344</td>\n",
       "      <td>0.060757</td>\n",
       "      <td>-0.015188</td>\n",
       "      <td>-0.262275</td>\n",
       "      <td>...</td>\n",
       "      <td>0.174712</td>\n",
       "      <td>-0.618977</td>\n",
       "      <td>0.086628</td>\n",
       "      <td>0.204789</td>\n",
       "      <td>0.563708</td>\n",
       "      <td>-0.575633</td>\n",
       "      <td>-0.712244</td>\n",
       "      <td>-0.512830</td>\n",
       "      <td>0.172632</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1.452229</td>\n",
       "      <td>-0.851816</td>\n",
       "      <td>0.276584</td>\n",
       "      <td>-0.014454</td>\n",
       "      <td>-0.943997</td>\n",
       "      <td>0.084479</td>\n",
       "      <td>-0.462501</td>\n",
       "      <td>0.000943</td>\n",
       "      <td>-0.261788</td>\n",
       "      <td>-0.238519</td>\n",
       "      <td>...</td>\n",
       "      <td>0.861395</td>\n",
       "      <td>0.574682</td>\n",
       "      <td>0.040226</td>\n",
       "      <td>-0.076251</td>\n",
       "      <td>-0.006449</td>\n",
       "      <td>0.509123</td>\n",
       "      <td>-0.832214</td>\n",
       "      <td>0.644480</td>\n",
       "      <td>-0.244227</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1.452193</td>\n",
       "      <td>1.316847</td>\n",
       "      <td>-0.665085</td>\n",
       "      <td>-0.609705</td>\n",
       "      <td>0.521875</td>\n",
       "      <td>1.240170</td>\n",
       "      <td>0.527401</td>\n",
       "      <td>0.652844</td>\n",
       "      <td>-1.184247</td>\n",
       "      <td>-0.266748</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.131785</td>\n",
       "      <td>0.715461</td>\n",
       "      <td>-0.006711</td>\n",
       "      <td>-0.178164</td>\n",
       "      <td>-0.391087</td>\n",
       "      <td>-0.788080</td>\n",
       "      <td>-0.790447</td>\n",
       "      <td>-2.198904</td>\n",
       "      <td>0.156094</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1.452157</td>\n",
       "      <td>1.079182</td>\n",
       "      <td>0.271939</td>\n",
       "      <td>-2.042608</td>\n",
       "      <td>-0.065764</td>\n",
       "      <td>0.171348</td>\n",
       "      <td>-0.446082</td>\n",
       "      <td>0.602331</td>\n",
       "      <td>-0.571258</td>\n",
       "      <td>-1.061541</td>\n",
       "      <td>...</td>\n",
       "      <td>0.333762</td>\n",
       "      <td>0.054319</td>\n",
       "      <td>0.478027</td>\n",
       "      <td>0.120796</td>\n",
       "      <td>-0.015046</td>\n",
       "      <td>-0.390516</td>\n",
       "      <td>-0.282131</td>\n",
       "      <td>-0.633253</td>\n",
       "      <td>0.263857</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>1.452157</td>\n",
       "      <td>-0.941475</td>\n",
       "      <td>0.285961</td>\n",
       "      <td>0.909578</td>\n",
       "      <td>1.089642</td>\n",
       "      <td>-0.537661</td>\n",
       "      <td>-0.129375</td>\n",
       "      <td>-0.410612</td>\n",
       "      <td>0.075865</td>\n",
       "      <td>-0.593604</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.024140</td>\n",
       "      <td>-1.717977</td>\n",
       "      <td>-0.326527</td>\n",
       "      <td>0.371249</td>\n",
       "      <td>1.402308</td>\n",
       "      <td>2.848965</td>\n",
       "      <td>-1.269394</td>\n",
       "      <td>1.930274</td>\n",
       "      <td>-0.243216</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>1.452121</td>\n",
       "      <td>-0.966556</td>\n",
       "      <td>0.738509</td>\n",
       "      <td>-0.506658</td>\n",
       "      <td>0.460218</td>\n",
       "      <td>1.157781</td>\n",
       "      <td>0.009149</td>\n",
       "      <td>1.020093</td>\n",
       "      <td>-0.193854</td>\n",
       "      <td>0.083470</td>\n",
       "      <td>...</td>\n",
       "      <td>0.251126</td>\n",
       "      <td>-1.056985</td>\n",
       "      <td>-0.218744</td>\n",
       "      <td>-0.517196</td>\n",
       "      <td>-1.237500</td>\n",
       "      <td>-0.034066</td>\n",
       "      <td>-0.700985</td>\n",
       "      <td>0.101164</td>\n",
       "      <td>-0.030156</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1.452085</td>\n",
       "      <td>1.398582</td>\n",
       "      <td>2.335082</td>\n",
       "      <td>-1.442586</td>\n",
       "      <td>0.384820</td>\n",
       "      <td>0.069173</td>\n",
       "      <td>0.697448</td>\n",
       "      <td>0.214838</td>\n",
       "      <td>-0.260732</td>\n",
       "      <td>0.474425</td>\n",
       "      <td>...</td>\n",
       "      <td>2.259725</td>\n",
       "      <td>0.324589</td>\n",
       "      <td>-2.491744</td>\n",
       "      <td>-0.776372</td>\n",
       "      <td>-0.440794</td>\n",
       "      <td>-0.957950</td>\n",
       "      <td>-1.206239</td>\n",
       "      <td>0.686962</td>\n",
       "      <td>-1.664437</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1.452048</td>\n",
       "      <td>0.743339</td>\n",
       "      <td>-0.612449</td>\n",
       "      <td>-1.047389</td>\n",
       "      <td>0.027507</td>\n",
       "      <td>-0.003666</td>\n",
       "      <td>0.474980</td>\n",
       "      <td>-0.587055</td>\n",
       "      <td>-0.029790</td>\n",
       "      <td>0.378340</td>\n",
       "      <td>...</td>\n",
       "      <td>1.095475</td>\n",
       "      <td>0.564961</td>\n",
       "      <td>-0.094992</td>\n",
       "      <td>0.219452</td>\n",
       "      <td>0.539188</td>\n",
       "      <td>-0.628537</td>\n",
       "      <td>0.666482</td>\n",
       "      <td>-0.162887</td>\n",
       "      <td>0.311855</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>1.452048</td>\n",
       "      <td>-0.948551</td>\n",
       "      <td>0.058835</td>\n",
       "      <td>0.368274</td>\n",
       "      <td>-0.100133</td>\n",
       "      <td>0.136339</td>\n",
       "      <td>1.057006</td>\n",
       "      <td>-0.548517</td>\n",
       "      <td>0.388016</td>\n",
       "      <td>0.210854</td>\n",
       "      <td>...</td>\n",
       "      <td>0.888399</td>\n",
       "      <td>-0.106383</td>\n",
       "      <td>-0.240624</td>\n",
       "      <td>-0.050274</td>\n",
       "      <td>0.128409</td>\n",
       "      <td>-0.829388</td>\n",
       "      <td>-0.981897</td>\n",
       "      <td>-2.180281</td>\n",
       "      <td>-0.122918</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1.452048</td>\n",
       "      <td>-0.908762</td>\n",
       "      <td>0.195007</td>\n",
       "      <td>-0.465370</td>\n",
       "      <td>-1.010256</td>\n",
       "      <td>0.466887</td>\n",
       "      <td>-0.160355</td>\n",
       "      <td>0.255458</td>\n",
       "      <td>-0.127634</td>\n",
       "      <td>-0.843811</td>\n",
       "      <td>...</td>\n",
       "      <td>0.805846</td>\n",
       "      <td>0.434133</td>\n",
       "      <td>0.076853</td>\n",
       "      <td>-0.050922</td>\n",
       "      <td>-0.363765</td>\n",
       "      <td>-0.170948</td>\n",
       "      <td>-1.046809</td>\n",
       "      <td>0.528577</td>\n",
       "      <td>0.010199</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>1.452012</td>\n",
       "      <td>-1.000698</td>\n",
       "      <td>-0.206222</td>\n",
       "      <td>0.236789</td>\n",
       "      <td>-0.185509</td>\n",
       "      <td>-0.288243</td>\n",
       "      <td>0.072148</td>\n",
       "      <td>-0.034873</td>\n",
       "      <td>0.028910</td>\n",
       "      <td>0.220763</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.592908</td>\n",
       "      <td>-0.651755</td>\n",
       "      <td>-0.011494</td>\n",
       "      <td>0.397043</td>\n",
       "      <td>1.180666</td>\n",
       "      <td>1.455129</td>\n",
       "      <td>-0.660974</td>\n",
       "      <td>-0.302093</td>\n",
       "      <td>0.354677</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>1.452012</td>\n",
       "      <td>-0.955679</td>\n",
       "      <td>0.614172</td>\n",
       "      <td>-0.499729</td>\n",
       "      <td>0.596064</td>\n",
       "      <td>0.808809</td>\n",
       "      <td>-0.404591</td>\n",
       "      <td>0.993609</td>\n",
       "      <td>-0.288186</td>\n",
       "      <td>0.569528</td>\n",
       "      <td>...</td>\n",
       "      <td>1.481136</td>\n",
       "      <td>0.270371</td>\n",
       "      <td>-0.183503</td>\n",
       "      <td>-0.721183</td>\n",
       "      <td>-1.964235</td>\n",
       "      <td>0.394169</td>\n",
       "      <td>-0.375218</td>\n",
       "      <td>0.030722</td>\n",
       "      <td>0.075261</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149970</th>\n",
       "      <td>-1.756428</td>\n",
       "      <td>1.016993</td>\n",
       "      <td>-1.300244</td>\n",
       "      <td>1.647278</td>\n",
       "      <td>0.570551</td>\n",
       "      <td>-0.350534</td>\n",
       "      <td>0.363522</td>\n",
       "      <td>0.061048</td>\n",
       "      <td>-1.010531</td>\n",
       "      <td>0.137970</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.172838</td>\n",
       "      <td>-0.095729</td>\n",
       "      <td>0.001744</td>\n",
       "      <td>0.442051</td>\n",
       "      <td>1.455620</td>\n",
       "      <td>0.699440</td>\n",
       "      <td>0.264657</td>\n",
       "      <td>-0.387327</td>\n",
       "      <td>0.321762</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149971</th>\n",
       "      <td>-1.756428</td>\n",
       "      <td>-1.043185</td>\n",
       "      <td>0.353700</td>\n",
       "      <td>1.752885</td>\n",
       "      <td>-0.701406</td>\n",
       "      <td>-1.040170</td>\n",
       "      <td>-0.820345</td>\n",
       "      <td>-0.539261</td>\n",
       "      <td>-0.166103</td>\n",
       "      <td>0.293354</td>\n",
       "      <td>...</td>\n",
       "      <td>1.990878</td>\n",
       "      <td>1.928310</td>\n",
       "      <td>0.038249</td>\n",
       "      <td>-0.540252</td>\n",
       "      <td>-1.263675</td>\n",
       "      <td>2.819244</td>\n",
       "      <td>-0.086130</td>\n",
       "      <td>0.778481</td>\n",
       "      <td>-0.486967</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149972</th>\n",
       "      <td>-1.756428</td>\n",
       "      <td>0.798683</td>\n",
       "      <td>0.362099</td>\n",
       "      <td>0.977508</td>\n",
       "      <td>0.021173</td>\n",
       "      <td>-0.275638</td>\n",
       "      <td>-0.050383</td>\n",
       "      <td>-0.553856</td>\n",
       "      <td>-0.111420</td>\n",
       "      <td>-0.445095</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.449245</td>\n",
       "      <td>-1.910909</td>\n",
       "      <td>0.235416</td>\n",
       "      <td>-0.143171</td>\n",
       "      <td>-0.673102</td>\n",
       "      <td>-0.537099</td>\n",
       "      <td>4.372295</td>\n",
       "      <td>0.697698</td>\n",
       "      <td>-0.424574</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149973</th>\n",
       "      <td>-1.756465</td>\n",
       "      <td>0.913511</td>\n",
       "      <td>-0.575903</td>\n",
       "      <td>0.139076</td>\n",
       "      <td>0.844986</td>\n",
       "      <td>-0.619400</td>\n",
       "      <td>0.592743</td>\n",
       "      <td>-0.788683</td>\n",
       "      <td>0.047656</td>\n",
       "      <td>0.262468</td>\n",
       "      <td>...</td>\n",
       "      <td>0.306497</td>\n",
       "      <td>0.170187</td>\n",
       "      <td>0.398145</td>\n",
       "      <td>0.105658</td>\n",
       "      <td>0.376347</td>\n",
       "      <td>0.541750</td>\n",
       "      <td>1.840479</td>\n",
       "      <td>0.199816</td>\n",
       "      <td>0.321924</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149974</th>\n",
       "      <td>-1.756465</td>\n",
       "      <td>0.753897</td>\n",
       "      <td>-0.179997</td>\n",
       "      <td>-1.012101</td>\n",
       "      <td>0.599390</td>\n",
       "      <td>-0.437494</td>\n",
       "      <td>-0.826563</td>\n",
       "      <td>-0.194584</td>\n",
       "      <td>-0.283565</td>\n",
       "      <td>-0.656740</td>\n",
       "      <td>...</td>\n",
       "      <td>1.471743</td>\n",
       "      <td>1.821583</td>\n",
       "      <td>0.032712</td>\n",
       "      <td>-0.322517</td>\n",
       "      <td>-1.683961</td>\n",
       "      <td>1.663371</td>\n",
       "      <td>1.254492</td>\n",
       "      <td>-1.287262</td>\n",
       "      <td>0.311774</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149975</th>\n",
       "      <td>-1.756501</td>\n",
       "      <td>0.496970</td>\n",
       "      <td>-0.225179</td>\n",
       "      <td>-0.534901</td>\n",
       "      <td>0.678979</td>\n",
       "      <td>-0.185976</td>\n",
       "      <td>0.008957</td>\n",
       "      <td>-0.360932</td>\n",
       "      <td>-0.150653</td>\n",
       "      <td>-0.042370</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.268163</td>\n",
       "      <td>-0.057454</td>\n",
       "      <td>0.084419</td>\n",
       "      <td>0.051480</td>\n",
       "      <td>0.280999</td>\n",
       "      <td>-1.227372</td>\n",
       "      <td>2.953577</td>\n",
       "      <td>-0.010524</td>\n",
       "      <td>0.209066</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149976</th>\n",
       "      <td>-1.756537</td>\n",
       "      <td>-0.466874</td>\n",
       "      <td>-0.586969</td>\n",
       "      <td>0.575110</td>\n",
       "      <td>0.585976</td>\n",
       "      <td>-0.949058</td>\n",
       "      <td>0.174812</td>\n",
       "      <td>-0.742782</td>\n",
       "      <td>-0.061101</td>\n",
       "      <td>0.200091</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.593735</td>\n",
       "      <td>-0.333932</td>\n",
       "      <td>-0.045221</td>\n",
       "      <td>0.426840</td>\n",
       "      <td>1.139958</td>\n",
       "      <td>0.038207</td>\n",
       "      <td>0.825265</td>\n",
       "      <td>-0.263748</td>\n",
       "      <td>0.334055</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149977</th>\n",
       "      <td>-1.756537</td>\n",
       "      <td>-1.159660</td>\n",
       "      <td>0.170874</td>\n",
       "      <td>0.917977</td>\n",
       "      <td>-0.387761</td>\n",
       "      <td>0.133238</td>\n",
       "      <td>0.688645</td>\n",
       "      <td>-0.023712</td>\n",
       "      <td>0.180386</td>\n",
       "      <td>-1.117878</td>\n",
       "      <td>...</td>\n",
       "      <td>0.394354</td>\n",
       "      <td>-0.159074</td>\n",
       "      <td>0.401454</td>\n",
       "      <td>-0.137221</td>\n",
       "      <td>-0.821826</td>\n",
       "      <td>0.108429</td>\n",
       "      <td>-0.405998</td>\n",
       "      <td>0.281043</td>\n",
       "      <td>0.302757</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149978</th>\n",
       "      <td>-1.756573</td>\n",
       "      <td>-1.059615</td>\n",
       "      <td>0.689100</td>\n",
       "      <td>1.290097</td>\n",
       "      <td>-0.255152</td>\n",
       "      <td>-1.236183</td>\n",
       "      <td>-3.189057</td>\n",
       "      <td>0.813775</td>\n",
       "      <td>-0.892983</td>\n",
       "      <td>-0.837991</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.048677</td>\n",
       "      <td>0.940568</td>\n",
       "      <td>-0.294563</td>\n",
       "      <td>-0.132336</td>\n",
       "      <td>0.185648</td>\n",
       "      <td>-1.083483</td>\n",
       "      <td>0.835051</td>\n",
       "      <td>1.706793</td>\n",
       "      <td>-0.471076</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149979</th>\n",
       "      <td>-1.756573</td>\n",
       "      <td>1.182746</td>\n",
       "      <td>0.966613</td>\n",
       "      <td>0.162129</td>\n",
       "      <td>1.441268</td>\n",
       "      <td>-1.106665</td>\n",
       "      <td>0.459592</td>\n",
       "      <td>-0.037243</td>\n",
       "      <td>0.008725</td>\n",
       "      <td>0.883619</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.029774</td>\n",
       "      <td>-0.780449</td>\n",
       "      <td>-0.240209</td>\n",
       "      <td>0.053925</td>\n",
       "      <td>0.095853</td>\n",
       "      <td>2.391011</td>\n",
       "      <td>-2.101367</td>\n",
       "      <td>0.152280</td>\n",
       "      <td>-0.573582</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149980</th>\n",
       "      <td>-1.756573</td>\n",
       "      <td>-0.391144</td>\n",
       "      <td>-0.612489</td>\n",
       "      <td>0.372026</td>\n",
       "      <td>0.482981</td>\n",
       "      <td>-0.761526</td>\n",
       "      <td>0.391464</td>\n",
       "      <td>-0.767989</td>\n",
       "      <td>-0.026102</td>\n",
       "      <td>0.304215</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.157774</td>\n",
       "      <td>0.062917</td>\n",
       "      <td>-0.046655</td>\n",
       "      <td>0.352909</td>\n",
       "      <td>0.881874</td>\n",
       "      <td>-1.003665</td>\n",
       "      <td>0.999803</td>\n",
       "      <td>-0.186210</td>\n",
       "      <td>0.334055</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149981</th>\n",
       "      <td>-1.756609</td>\n",
       "      <td>-1.147584</td>\n",
       "      <td>0.279278</td>\n",
       "      <td>0.281969</td>\n",
       "      <td>-0.345473</td>\n",
       "      <td>0.565136</td>\n",
       "      <td>0.462859</td>\n",
       "      <td>0.455514</td>\n",
       "      <td>0.103554</td>\n",
       "      <td>-1.049987</td>\n",
       "      <td>...</td>\n",
       "      <td>0.054809</td>\n",
       "      <td>0.372779</td>\n",
       "      <td>0.135234</td>\n",
       "      <td>-0.338354</td>\n",
       "      <td>-1.350828</td>\n",
       "      <td>-0.223581</td>\n",
       "      <td>0.482992</td>\n",
       "      <td>-1.278385</td>\n",
       "      <td>0.237007</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149982</th>\n",
       "      <td>-1.756645</td>\n",
       "      <td>1.165711</td>\n",
       "      <td>-1.533038</td>\n",
       "      <td>1.598313</td>\n",
       "      <td>0.394605</td>\n",
       "      <td>0.319690</td>\n",
       "      <td>0.755502</td>\n",
       "      <td>0.317364</td>\n",
       "      <td>-1.255737</td>\n",
       "      <td>0.179836</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.196970</td>\n",
       "      <td>0.130018</td>\n",
       "      <td>0.032535</td>\n",
       "      <td>-0.621675</td>\n",
       "      <td>-1.788662</td>\n",
       "      <td>-1.183867</td>\n",
       "      <td>1.078993</td>\n",
       "      <td>0.471305</td>\n",
       "      <td>0.323946</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149983</th>\n",
       "      <td>-1.756681</td>\n",
       "      <td>-1.119734</td>\n",
       "      <td>0.048303</td>\n",
       "      <td>1.680693</td>\n",
       "      <td>-0.843280</td>\n",
       "      <td>-0.628014</td>\n",
       "      <td>0.686008</td>\n",
       "      <td>-0.810070</td>\n",
       "      <td>0.374889</td>\n",
       "      <td>0.138759</td>\n",
       "      <td>...</td>\n",
       "      <td>0.479067</td>\n",
       "      <td>0.527980</td>\n",
       "      <td>-0.024970</td>\n",
       "      <td>-0.281208</td>\n",
       "      <td>-0.521394</td>\n",
       "      <td>-1.005216</td>\n",
       "      <td>-0.810031</td>\n",
       "      <td>1.082404</td>\n",
       "      <td>-0.224009</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149984</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.449562</td>\n",
       "      <td>-0.193735</td>\n",
       "      <td>-0.449793</td>\n",
       "      <td>-0.362727</td>\n",
       "      <td>0.273393</td>\n",
       "      <td>-1.050009</td>\n",
       "      <td>-0.940083</td>\n",
       "      <td>0.474071</td>\n",
       "      <td>-0.620439</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.929034</td>\n",
       "      <td>-3.555896</td>\n",
       "      <td>-0.419077</td>\n",
       "      <td>-0.187227</td>\n",
       "      <td>-1.684362</td>\n",
       "      <td>-0.135998</td>\n",
       "      <td>1.753141</td>\n",
       "      <td>-0.855840</td>\n",
       "      <td>-0.450453</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149985</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.382955</td>\n",
       "      <td>-0.643444</td>\n",
       "      <td>0.903562</td>\n",
       "      <td>0.007353</td>\n",
       "      <td>-0.176866</td>\n",
       "      <td>1.013712</td>\n",
       "      <td>-0.805823</td>\n",
       "      <td>0.000068</td>\n",
       "      <td>0.325499</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.717082</td>\n",
       "      <td>0.334866</td>\n",
       "      <td>0.006529</td>\n",
       "      <td>-0.538168</td>\n",
       "      <td>-1.524937</td>\n",
       "      <td>0.004144</td>\n",
       "      <td>0.454743</td>\n",
       "      <td>0.331630</td>\n",
       "      <td>0.013555</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149986</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.964570</td>\n",
       "      <td>1.117810</td>\n",
       "      <td>0.239642</td>\n",
       "      <td>-0.738516</td>\n",
       "      <td>0.994184</td>\n",
       "      <td>-0.236225</td>\n",
       "      <td>0.394180</td>\n",
       "      <td>-0.140536</td>\n",
       "      <td>-1.340730</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.006939</td>\n",
       "      <td>-0.311363</td>\n",
       "      <td>-0.678934</td>\n",
       "      <td>-0.133608</td>\n",
       "      <td>0.306317</td>\n",
       "      <td>-0.079563</td>\n",
       "      <td>1.341826</td>\n",
       "      <td>-0.126860</td>\n",
       "      <td>-1.173457</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149987</th>\n",
       "      <td>-1.756717</td>\n",
       "      <td>-0.513342</td>\n",
       "      <td>-0.677014</td>\n",
       "      <td>0.863835</td>\n",
       "      <td>0.024536</td>\n",
       "      <td>-0.201272</td>\n",
       "      <td>1.122639</td>\n",
       "      <td>-0.526677</td>\n",
       "      <td>-0.095007</td>\n",
       "      <td>0.076104</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.931695</td>\n",
       "      <td>0.301890</td>\n",
       "      <td>0.329482</td>\n",
       "      <td>-0.461505</td>\n",
       "      <td>-1.339734</td>\n",
       "      <td>0.066709</td>\n",
       "      <td>0.617299</td>\n",
       "      <td>0.315958</td>\n",
       "      <td>0.272227</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149988</th>\n",
       "      <td>-1.756753</td>\n",
       "      <td>1.019253</td>\n",
       "      <td>-0.875585</td>\n",
       "      <td>0.468502</td>\n",
       "      <td>1.310742</td>\n",
       "      <td>-0.921393</td>\n",
       "      <td>0.088205</td>\n",
       "      <td>-0.842800</td>\n",
       "      <td>-0.111500</td>\n",
       "      <td>0.088516</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.147286</td>\n",
       "      <td>0.287595</td>\n",
       "      <td>0.525987</td>\n",
       "      <td>0.046710</td>\n",
       "      <td>0.536748</td>\n",
       "      <td>2.323463</td>\n",
       "      <td>0.154244</td>\n",
       "      <td>1.656127</td>\n",
       "      <td>0.310642</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149989</th>\n",
       "      <td>-1.756753</td>\n",
       "      <td>-1.128135</td>\n",
       "      <td>0.241159</td>\n",
       "      <td>0.848114</td>\n",
       "      <td>-0.346470</td>\n",
       "      <td>0.184799</td>\n",
       "      <td>0.745318</td>\n",
       "      <td>-0.136615</td>\n",
       "      <td>0.250803</td>\n",
       "      <td>-0.799592</td>\n",
       "      <td>...</td>\n",
       "      <td>0.457895</td>\n",
       "      <td>0.150482</td>\n",
       "      <td>0.066831</td>\n",
       "      <td>-0.111768</td>\n",
       "      <td>-0.377428</td>\n",
       "      <td>-0.002900</td>\n",
       "      <td>0.118576</td>\n",
       "      <td>0.721861</td>\n",
       "      <td>-0.020411</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149990</th>\n",
       "      <td>-1.756789</td>\n",
       "      <td>0.720982</td>\n",
       "      <td>-0.697570</td>\n",
       "      <td>-0.215227</td>\n",
       "      <td>-0.409272</td>\n",
       "      <td>-0.783372</td>\n",
       "      <td>0.368362</td>\n",
       "      <td>-1.490595</td>\n",
       "      <td>0.258694</td>\n",
       "      <td>0.989876</td>\n",
       "      <td>...</td>\n",
       "      <td>0.801457</td>\n",
       "      <td>-0.090667</td>\n",
       "      <td>-0.160423</td>\n",
       "      <td>-0.146275</td>\n",
       "      <td>-0.474964</td>\n",
       "      <td>-1.858457</td>\n",
       "      <td>-2.557826</td>\n",
       "      <td>0.708178</td>\n",
       "      <td>0.067942</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149991</th>\n",
       "      <td>-1.756789</td>\n",
       "      <td>0.844073</td>\n",
       "      <td>-0.297288</td>\n",
       "      <td>-0.037355</td>\n",
       "      <td>-0.909128</td>\n",
       "      <td>0.373785</td>\n",
       "      <td>-0.638166</td>\n",
       "      <td>-0.733955</td>\n",
       "      <td>-0.311541</td>\n",
       "      <td>-0.060952</td>\n",
       "      <td>...</td>\n",
       "      <td>0.789836</td>\n",
       "      <td>-1.363126</td>\n",
       "      <td>0.122533</td>\n",
       "      <td>0.512972</td>\n",
       "      <td>1.338080</td>\n",
       "      <td>-0.914364</td>\n",
       "      <td>1.219112</td>\n",
       "      <td>2.135645</td>\n",
       "      <td>-0.532013</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149992</th>\n",
       "      <td>-1.756789</td>\n",
       "      <td>-1.158606</td>\n",
       "      <td>0.080272</td>\n",
       "      <td>1.628651</td>\n",
       "      <td>-0.077196</td>\n",
       "      <td>-0.490708</td>\n",
       "      <td>0.693930</td>\n",
       "      <td>-0.519585</td>\n",
       "      <td>0.275900</td>\n",
       "      <td>0.047585</td>\n",
       "      <td>...</td>\n",
       "      <td>0.300039</td>\n",
       "      <td>-0.543964</td>\n",
       "      <td>0.170318</td>\n",
       "      <td>-0.239883</td>\n",
       "      <td>-0.763969</td>\n",
       "      <td>-1.408738</td>\n",
       "      <td>-0.556032</td>\n",
       "      <td>-1.583964</td>\n",
       "      <td>0.174856</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149993</th>\n",
       "      <td>-1.756825</td>\n",
       "      <td>-1.163653</td>\n",
       "      <td>0.253798</td>\n",
       "      <td>2.003056</td>\n",
       "      <td>0.513224</td>\n",
       "      <td>-1.844906</td>\n",
       "      <td>-2.527033</td>\n",
       "      <td>0.338701</td>\n",
       "      <td>-0.642826</td>\n",
       "      <td>-0.424007</td>\n",
       "      <td>...</td>\n",
       "      <td>1.192495</td>\n",
       "      <td>-0.107367</td>\n",
       "      <td>0.196511</td>\n",
       "      <td>0.321263</td>\n",
       "      <td>0.925534</td>\n",
       "      <td>-1.177971</td>\n",
       "      <td>0.392388</td>\n",
       "      <td>-0.610395</td>\n",
       "      <td>0.310440</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149994</th>\n",
       "      <td>-1.756825</td>\n",
       "      <td>-1.165644</td>\n",
       "      <td>0.935357</td>\n",
       "      <td>1.386949</td>\n",
       "      <td>1.325895</td>\n",
       "      <td>-0.858063</td>\n",
       "      <td>-2.924312</td>\n",
       "      <td>1.457903</td>\n",
       "      <td>-0.835015</td>\n",
       "      <td>-0.189523</td>\n",
       "      <td>...</td>\n",
       "      <td>1.317563</td>\n",
       "      <td>-0.321494</td>\n",
       "      <td>-0.180986</td>\n",
       "      <td>-0.633114</td>\n",
       "      <td>-1.798527</td>\n",
       "      <td>-1.261384</td>\n",
       "      <td>0.239633</td>\n",
       "      <td>-0.079566</td>\n",
       "      <td>0.116547</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149995</th>\n",
       "      <td>-1.756898</td>\n",
       "      <td>0.867969</td>\n",
       "      <td>-0.905587</td>\n",
       "      <td>1.055569</td>\n",
       "      <td>0.517767</td>\n",
       "      <td>-0.276563</td>\n",
       "      <td>0.266143</td>\n",
       "      <td>-0.074745</td>\n",
       "      <td>-0.774578</td>\n",
       "      <td>0.146479</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.209218</td>\n",
       "      <td>-0.511090</td>\n",
       "      <td>0.222086</td>\n",
       "      <td>-0.592917</td>\n",
       "      <td>-1.703160</td>\n",
       "      <td>-0.347522</td>\n",
       "      <td>1.462083</td>\n",
       "      <td>0.469627</td>\n",
       "      <td>0.354273</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149996</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>0.415955</td>\n",
       "      <td>-1.200718</td>\n",
       "      <td>-0.259352</td>\n",
       "      <td>-2.851478</td>\n",
       "      <td>-0.421152</td>\n",
       "      <td>-0.370122</td>\n",
       "      <td>-2.084808</td>\n",
       "      <td>1.104757</td>\n",
       "      <td>0.490121</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.078761</td>\n",
       "      <td>-0.924988</td>\n",
       "      <td>-1.257517</td>\n",
       "      <td>0.084737</td>\n",
       "      <td>-1.981259</td>\n",
       "      <td>-0.110895</td>\n",
       "      <td>2.360976</td>\n",
       "      <td>-0.549872</td>\n",
       "      <td>-0.254902</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149997</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>1.168238</td>\n",
       "      <td>-0.552907</td>\n",
       "      <td>0.423081</td>\n",
       "      <td>0.155091</td>\n",
       "      <td>-0.270602</td>\n",
       "      <td>0.121991</td>\n",
       "      <td>-1.100427</td>\n",
       "      <td>0.400494</td>\n",
       "      <td>-0.304126</td>\n",
       "      <td>...</td>\n",
       "      <td>0.125388</td>\n",
       "      <td>-2.101888</td>\n",
       "      <td>0.433053</td>\n",
       "      <td>0.448856</td>\n",
       "      <td>0.183284</td>\n",
       "      <td>0.718502</td>\n",
       "      <td>1.235442</td>\n",
       "      <td>-0.582444</td>\n",
       "      <td>-0.079287</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149998</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>0.969417</td>\n",
       "      <td>-0.579037</td>\n",
       "      <td>-1.038397</td>\n",
       "      <td>0.288642</td>\n",
       "      <td>0.177344</td>\n",
       "      <td>-0.353638</td>\n",
       "      <td>0.066474</td>\n",
       "      <td>-0.686863</td>\n",
       "      <td>-0.341261</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.279584</td>\n",
       "      <td>-1.593140</td>\n",
       "      <td>0.088111</td>\n",
       "      <td>0.470687</td>\n",
       "      <td>1.151563</td>\n",
       "      <td>0.863148</td>\n",
       "      <td>-0.831973</td>\n",
       "      <td>0.858601</td>\n",
       "      <td>0.354233</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149999</th>\n",
       "      <td>-1.756934</td>\n",
       "      <td>-1.217029</td>\n",
       "      <td>0.306214</td>\n",
       "      <td>1.981126</td>\n",
       "      <td>0.715126</td>\n",
       "      <td>-0.286982</td>\n",
       "      <td>1.128890</td>\n",
       "      <td>-0.298423</td>\n",
       "      <td>0.434975</td>\n",
       "      <td>0.752355</td>\n",
       "      <td>...</td>\n",
       "      <td>1.961673</td>\n",
       "      <td>-0.893321</td>\n",
       "      <td>0.218858</td>\n",
       "      <td>-0.625348</td>\n",
       "      <td>-1.911075</td>\n",
       "      <td>-1.219105</td>\n",
       "      <td>-1.381696</td>\n",
       "      <td>-0.572706</td>\n",
       "      <td>0.351038</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150000 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Ocorrencia       PP1       PP2       PP3       PP4       PP5  \\\n",
       "0         1.452554 -0.985473  0.607243 -0.578142  0.206595  0.604854   \n",
       "1         1.452518 -0.711549  1.151666 -0.060850  0.830176  1.091942   \n",
       "2         1.452481  0.615691  2.128527 -0.377129 -0.279669  1.826966   \n",
       "3         1.452481 -0.935247 -0.064476 -0.024557 -0.638310 -0.120940   \n",
       "4         1.452445  1.555849  1.615105  0.756921  0.294379  1.039644   \n",
       "5         1.452445 -0.981952 -0.034877 -0.322386 -0.838697  0.332925   \n",
       "6         1.452445 -0.971092 -0.174094  0.022727 -0.596579 -0.163207   \n",
       "7         1.452445  1.423358  2.943124 -0.502896  0.271327 -2.055703   \n",
       "8         1.452445 -0.994288  0.265920 -0.129913  1.049748  0.553487   \n",
       "9         1.452409  0.893649  0.262019 -0.938725  1.166103  0.781218   \n",
       "10        1.452373  0.910005 -0.475790 -1.084986  0.465173 -0.158645   \n",
       "11        1.452373 -0.951713  0.070342 -0.817547 -0.840311  0.795529   \n",
       "12        1.452337  1.226187  0.626067 -1.426541 -0.249718 -1.193089   \n",
       "13        1.452301 -1.045434  0.373695  0.080851  0.870383  0.242749   \n",
       "14        1.452265 -1.007026  0.585796  1.033456  1.288880 -1.089464   \n",
       "15        1.452265 -1.013460  0.539064 -0.484548  1.452957  1.318202   \n",
       "16        1.452265 -0.978759 -0.025438  0.312878 -0.527179 -0.166151   \n",
       "17        1.452229  0.844634 -0.591753 -0.085803 -1.356474 -1.318614   \n",
       "18        1.452229 -0.950305  0.076973 -0.223885 -0.518524  0.382386   \n",
       "19        1.452229 -0.851816  0.276584 -0.014454 -0.943997  0.084479   \n",
       "20        1.452193  1.316847 -0.665085 -0.609705  0.521875  1.240170   \n",
       "21        1.452157  1.079182  0.271939 -2.042608 -0.065764  0.171348   \n",
       "22        1.452157 -0.941475  0.285961  0.909578  1.089642 -0.537661   \n",
       "23        1.452121 -0.966556  0.738509 -0.506658  0.460218  1.157781   \n",
       "24        1.452085  1.398582  2.335082 -1.442586  0.384820  0.069173   \n",
       "25        1.452048  0.743339 -0.612449 -1.047389  0.027507 -0.003666   \n",
       "26        1.452048 -0.948551  0.058835  0.368274 -0.100133  0.136339   \n",
       "27        1.452048 -0.908762  0.195007 -0.465370 -1.010256  0.466887   \n",
       "28        1.452012 -1.000698 -0.206222  0.236789 -0.185509 -0.288243   \n",
       "29        1.452012 -0.955679  0.614172 -0.499729  0.596064  0.808809   \n",
       "...            ...       ...       ...       ...       ...       ...   \n",
       "149970   -1.756428  1.016993 -1.300244  1.647278  0.570551 -0.350534   \n",
       "149971   -1.756428 -1.043185  0.353700  1.752885 -0.701406 -1.040170   \n",
       "149972   -1.756428  0.798683  0.362099  0.977508  0.021173 -0.275638   \n",
       "149973   -1.756465  0.913511 -0.575903  0.139076  0.844986 -0.619400   \n",
       "149974   -1.756465  0.753897 -0.179997 -1.012101  0.599390 -0.437494   \n",
       "149975   -1.756501  0.496970 -0.225179 -0.534901  0.678979 -0.185976   \n",
       "149976   -1.756537 -0.466874 -0.586969  0.575110  0.585976 -0.949058   \n",
       "149977   -1.756537 -1.159660  0.170874  0.917977 -0.387761  0.133238   \n",
       "149978   -1.756573 -1.059615  0.689100  1.290097 -0.255152 -1.236183   \n",
       "149979   -1.756573  1.182746  0.966613  0.162129  1.441268 -1.106665   \n",
       "149980   -1.756573 -0.391144 -0.612489  0.372026  0.482981 -0.761526   \n",
       "149981   -1.756609 -1.147584  0.279278  0.281969 -0.345473  0.565136   \n",
       "149982   -1.756645  1.165711 -1.533038  1.598313  0.394605  0.319690   \n",
       "149983   -1.756681 -1.119734  0.048303  1.680693 -0.843280 -0.628014   \n",
       "149984   -1.756717 -0.449562 -0.193735 -0.449793 -0.362727  0.273393   \n",
       "149985   -1.756717 -0.382955 -0.643444  0.903562  0.007353 -0.176866   \n",
       "149986   -1.756717 -0.964570  1.117810  0.239642 -0.738516  0.994184   \n",
       "149987   -1.756717 -0.513342 -0.677014  0.863835  0.024536 -0.201272   \n",
       "149988   -1.756753  1.019253 -0.875585  0.468502  1.310742 -0.921393   \n",
       "149989   -1.756753 -1.128135  0.241159  0.848114 -0.346470  0.184799   \n",
       "149990   -1.756789  0.720982 -0.697570 -0.215227 -0.409272 -0.783372   \n",
       "149991   -1.756789  0.844073 -0.297288 -0.037355 -0.909128  0.373785   \n",
       "149992   -1.756789 -1.158606  0.080272  1.628651 -0.077196 -0.490708   \n",
       "149993   -1.756825 -1.163653  0.253798  2.003056  0.513224 -1.844906   \n",
       "149994   -1.756825 -1.165644  0.935357  1.386949  1.325895 -0.858063   \n",
       "149995   -1.756898  0.867969 -0.905587  1.055569  0.517767 -0.276563   \n",
       "149996   -1.756934  0.415955 -1.200718 -0.259352 -2.851478 -0.421152   \n",
       "149997   -1.756934  1.168238 -0.552907  0.423081  0.155091 -0.270602   \n",
       "149998   -1.756934  0.969417 -0.579037 -1.038397  0.288642  0.177344   \n",
       "149999   -1.756934 -1.217029  0.306214  1.981126  0.715126 -0.286982   \n",
       "\n",
       "             PP6       PP7       PP8       PP9   ...        PP18      PP19  \\\n",
       "0      -1.260358  1.192228 -0.393617 -0.033998   ...   -1.785697  0.216948   \n",
       "1       0.219741  0.180462 -0.001535 -1.525840   ...   -1.085263 -1.182226   \n",
       "2      -0.392920  0.266475 -0.007794 -0.236547   ...    2.331057 -0.279733   \n",
       "3      -0.209379  0.002692 -0.167246  0.185366   ...    0.600185  0.981217   \n",
       "4      -1.149025 -0.158615 -1.100691 -0.333398   ...   -0.319784 -0.443985   \n",
       "5       0.196228  0.192049 -0.057406 -0.753223   ...    0.265489 -0.316328   \n",
       "6       0.004658 -0.057781 -0.047207  0.239018   ...    0.013302  0.376667   \n",
       "7       0.641306  1.064781 -0.404160  1.005183   ...   -2.141295  1.232680   \n",
       "8       0.475634  0.293952 -0.014615 -1.404025   ...   -0.079015 -1.945638   \n",
       "9       0.405424 -0.717286  0.049532  1.476413   ...   -2.134925  1.511254   \n",
       "10      0.098420 -0.481386 -0.070508 -0.015903   ...   -0.162083  0.000508   \n",
       "11      0.121590  0.601885 -0.231436 -0.664958   ...   -0.838740  0.160756   \n",
       "12     -0.654438  1.212763 -0.530135 -0.808822   ...    2.410552  1.512248   \n",
       "13     -0.288918  0.727344 -0.101179  0.657392   ...    0.312318 -1.743823   \n",
       "14     -2.431620  0.794830 -0.612916  0.824701   ...    1.105854 -1.066890   \n",
       "15      1.009762  0.684201  0.174900 -0.417668   ...   -1.629926  0.420063   \n",
       "16      0.023364 -0.034418 -0.064410 -0.090317   ...   -1.171218 -0.449572   \n",
       "17     -3.558593  0.614877 -1.267491  0.902074   ...   -0.892266 -1.582167   \n",
       "18      0.316344  0.060757 -0.015188 -0.262275   ...    0.174712 -0.618977   \n",
       "19     -0.462501  0.000943 -0.261788 -0.238519   ...    0.861395  0.574682   \n",
       "20      0.527401  0.652844 -1.184247 -0.266748   ...   -0.131785  0.715461   \n",
       "21     -0.446082  0.602331 -0.571258 -1.061541   ...    0.333762  0.054319   \n",
       "22     -0.129375 -0.410612  0.075865 -0.593604   ...   -0.024140 -1.717977   \n",
       "23      0.009149  1.020093 -0.193854  0.083470   ...    0.251126 -1.056985   \n",
       "24      0.697448  0.214838 -0.260732  0.474425   ...    2.259725  0.324589   \n",
       "25      0.474980 -0.587055 -0.029790  0.378340   ...    1.095475  0.564961   \n",
       "26      1.057006 -0.548517  0.388016  0.210854   ...    0.888399 -0.106383   \n",
       "27     -0.160355  0.255458 -0.127634 -0.843811   ...    0.805846  0.434133   \n",
       "28      0.072148 -0.034873  0.028910  0.220763   ...   -0.592908 -0.651755   \n",
       "29     -0.404591  0.993609 -0.288186  0.569528   ...    1.481136  0.270371   \n",
       "...          ...       ...       ...       ...   ...         ...       ...   \n",
       "149970  0.363522  0.061048 -1.010531  0.137970   ...   -0.172838 -0.095729   \n",
       "149971 -0.820345 -0.539261 -0.166103  0.293354   ...    1.990878  1.928310   \n",
       "149972 -0.050383 -0.553856 -0.111420 -0.445095   ...   -0.449245 -1.910909   \n",
       "149973  0.592743 -0.788683  0.047656  0.262468   ...    0.306497  0.170187   \n",
       "149974 -0.826563 -0.194584 -0.283565 -0.656740   ...    1.471743  1.821583   \n",
       "149975  0.008957 -0.360932 -0.150653 -0.042370   ...   -0.268163 -0.057454   \n",
       "149976  0.174812 -0.742782 -0.061101  0.200091   ...   -0.593735 -0.333932   \n",
       "149977  0.688645 -0.023712  0.180386 -1.117878   ...    0.394354 -0.159074   \n",
       "149978 -3.189057  0.813775 -0.892983 -0.837991   ...   -0.048677  0.940568   \n",
       "149979  0.459592 -0.037243  0.008725  0.883619   ...   -0.029774 -0.780449   \n",
       "149980  0.391464 -0.767989 -0.026102  0.304215   ...   -0.157774  0.062917   \n",
       "149981  0.462859  0.455514  0.103554 -1.049987   ...    0.054809  0.372779   \n",
       "149982  0.755502  0.317364 -1.255737  0.179836   ...   -0.196970  0.130018   \n",
       "149983  0.686008 -0.810070  0.374889  0.138759   ...    0.479067  0.527980   \n",
       "149984 -1.050009 -0.940083  0.474071 -0.620439   ...   -0.929034 -3.555896   \n",
       "149985  1.013712 -0.805823  0.000068  0.325499   ...   -0.717082  0.334866   \n",
       "149986 -0.236225  0.394180 -0.140536 -1.340730   ...   -0.006939 -0.311363   \n",
       "149987  1.122639 -0.526677 -0.095007  0.076104   ...   -0.931695  0.301890   \n",
       "149988  0.088205 -0.842800 -0.111500  0.088516   ...   -0.147286  0.287595   \n",
       "149989  0.745318 -0.136615  0.250803 -0.799592   ...    0.457895  0.150482   \n",
       "149990  0.368362 -1.490595  0.258694  0.989876   ...    0.801457 -0.090667   \n",
       "149991 -0.638166 -0.733955 -0.311541 -0.060952   ...    0.789836 -1.363126   \n",
       "149992  0.693930 -0.519585  0.275900  0.047585   ...    0.300039 -0.543964   \n",
       "149993 -2.527033  0.338701 -0.642826 -0.424007   ...    1.192495 -0.107367   \n",
       "149994 -2.924312  1.457903 -0.835015 -0.189523   ...    1.317563 -0.321494   \n",
       "149995  0.266143 -0.074745 -0.774578  0.146479   ...   -1.209218 -0.511090   \n",
       "149996 -0.370122 -2.084808  1.104757  0.490121   ...   -0.078761 -0.924988   \n",
       "149997  0.121991 -1.100427  0.400494 -0.304126   ...    0.125388 -2.101888   \n",
       "149998 -0.353638  0.066474 -0.686863 -0.341261   ...   -0.279584 -1.593140   \n",
       "149999  1.128890 -0.298423  0.434975  0.752355   ...    1.961673 -0.893321   \n",
       "\n",
       "            PP20      PP21      PP22      PP24      PP25      PP26    Sacado  \\\n",
       "0       0.503995  0.510952  0.365689  2.141398 -0.525371 -1.389421  0.243518   \n",
       "1      -0.984745 -0.537851 -0.652846 -0.174406 -0.623881 -0.150271 -1.287488   \n",
       "2      -2.183033 -0.865822 -0.587094 -0.860282 -0.068399  0.408465 -2.876641   \n",
       "3       0.230558 -0.212839 -0.641821  0.481502 -0.894097  0.598090  0.231791   \n",
       "4       0.991586  0.285680  1.097367  2.515379  0.583625 -1.778647 -1.753154   \n",
       "5       0.316037  0.516861  1.400395  0.326806 -0.708302  1.250213  0.349097   \n",
       "6       0.203881 -0.129952 -0.481487  0.492480 -1.189520  0.614738  0.352211   \n",
       "7      -1.737285 -0.333678  0.649857  2.315159  0.180772 -0.699595 -1.198487   \n",
       "8       0.137952  0.426987  0.975534 -0.031139 -0.642520  0.584208  0.353666   \n",
       "9       0.087955 -0.036519 -0.132637 -0.909979  0.154585 -2.006428 -0.464282   \n",
       "10      0.256279 -0.071417 -0.223865 -0.090307 -0.457207 -0.748568  0.344973   \n",
       "11      0.316081 -0.065126 -0.275068 -0.805109 -0.495667  0.879377  0.353505   \n",
       "12      1.377398  0.110948 -0.758486  1.457480  1.576930 -0.812224  0.285167   \n",
       "13     -0.165211  0.015303  0.215884  2.332584 -0.826810  0.574471  0.333327   \n",
       "14     -0.451001 -0.201954 -0.053102 -1.702112 -1.178848  0.417280  0.007530   \n",
       "15      0.557118  0.412909  0.419496 -1.190853 -0.262500  0.200119  0.276595   \n",
       "16      0.196602 -0.017233  0.161675  1.463277 -1.391687  0.562756  0.199927   \n",
       "17     -0.146567 -0.271966 -0.658121 -1.727251  0.422321 -0.686513  0.170368   \n",
       "18      0.086628  0.204789  0.563708 -0.575633 -0.712244 -0.512830  0.172632   \n",
       "19      0.040226 -0.076251 -0.006449  0.509123 -0.832214  0.644480 -0.244227   \n",
       "20     -0.006711 -0.178164 -0.391087 -0.788080 -0.790447 -2.198904  0.156094   \n",
       "21      0.478027  0.120796 -0.015046 -0.390516 -0.282131 -0.633253  0.263857   \n",
       "22     -0.326527  0.371249  1.402308  2.848965 -1.269394  1.930274 -0.243216   \n",
       "23     -0.218744 -0.517196 -1.237500 -0.034066 -0.700985  0.101164 -0.030156   \n",
       "24     -2.491744 -0.776372 -0.440794 -0.957950 -1.206239  0.686962 -1.664437   \n",
       "25     -0.094992  0.219452  0.539188 -0.628537  0.666482 -0.162887  0.311855   \n",
       "26     -0.240624 -0.050274  0.128409 -0.829388 -0.981897 -2.180281 -0.122918   \n",
       "27      0.076853 -0.050922 -0.363765 -0.170948 -1.046809  0.528577  0.010199   \n",
       "28     -0.011494  0.397043  1.180666  1.455129 -0.660974 -0.302093  0.354677   \n",
       "29     -0.183503 -0.721183 -1.964235  0.394169 -0.375218  0.030722  0.075261   \n",
       "...          ...       ...       ...       ...       ...       ...       ...   \n",
       "149970  0.001744  0.442051  1.455620  0.699440  0.264657 -0.387327  0.321762   \n",
       "149971  0.038249 -0.540252 -1.263675  2.819244 -0.086130  0.778481 -0.486967   \n",
       "149972  0.235416 -0.143171 -0.673102 -0.537099  4.372295  0.697698 -0.424574   \n",
       "149973  0.398145  0.105658  0.376347  0.541750  1.840479  0.199816  0.321924   \n",
       "149974  0.032712 -0.322517 -1.683961  1.663371  1.254492 -1.287262  0.311774   \n",
       "149975  0.084419  0.051480  0.280999 -1.227372  2.953577 -0.010524  0.209066   \n",
       "149976 -0.045221  0.426840  1.139958  0.038207  0.825265 -0.263748  0.334055   \n",
       "149977  0.401454 -0.137221 -0.821826  0.108429 -0.405998  0.281043  0.302757   \n",
       "149978 -0.294563 -0.132336  0.185648 -1.083483  0.835051  1.706793 -0.471076   \n",
       "149979 -0.240209  0.053925  0.095853  2.391011 -2.101367  0.152280 -0.573582   \n",
       "149980 -0.046655  0.352909  0.881874 -1.003665  0.999803 -0.186210  0.334055   \n",
       "149981  0.135234 -0.338354 -1.350828 -0.223581  0.482992 -1.278385  0.237007   \n",
       "149982  0.032535 -0.621675 -1.788662 -1.183867  1.078993  0.471305  0.323946   \n",
       "149983 -0.024970 -0.281208 -0.521394 -1.005216 -0.810031  1.082404 -0.224009   \n",
       "149984 -0.419077 -0.187227 -1.684362 -0.135998  1.753141 -0.855840 -0.450453   \n",
       "149985  0.006529 -0.538168 -1.524937  0.004144  0.454743  0.331630  0.013555   \n",
       "149986 -0.678934 -0.133608  0.306317 -0.079563  1.341826 -0.126860 -1.173457   \n",
       "149987  0.329482 -0.461505 -1.339734  0.066709  0.617299  0.315958  0.272227   \n",
       "149988  0.525987  0.046710  0.536748  2.323463  0.154244  1.656127  0.310642   \n",
       "149989  0.066831 -0.111768 -0.377428 -0.002900  0.118576  0.721861 -0.020411   \n",
       "149990 -0.160423 -0.146275 -0.474964 -1.858457 -2.557826  0.708178  0.067942   \n",
       "149991  0.122533  0.512972  1.338080 -0.914364  1.219112  2.135645 -0.532013   \n",
       "149992  0.170318 -0.239883 -0.763969 -1.408738 -0.556032 -1.583964  0.174856   \n",
       "149993  0.196511  0.321263  0.925534 -1.177971  0.392388 -0.610395  0.310440   \n",
       "149994 -0.180986 -0.633114 -1.798527 -1.261384  0.239633 -0.079566  0.116547   \n",
       "149995  0.222086 -0.592917 -1.703160 -0.347522  1.462083  0.469627  0.354273   \n",
       "149996 -1.257517  0.084737 -1.981259 -0.110895  2.360976 -0.549872 -0.254902   \n",
       "149997  0.433053  0.448856  0.183284  0.718502  1.235442 -0.582444 -0.079287   \n",
       "149998  0.088111  0.470687  1.151563  0.863148 -0.831973  0.858601  0.354233   \n",
       "149999  0.218858 -0.625348 -1.911075 -1.219105 -1.381696 -0.572706  0.351038   \n",
       "\n",
       "        Fraude  \n",
       "0            0  \n",
       "1            0  \n",
       "2            0  \n",
       "3            0  \n",
       "4            0  \n",
       "5            0  \n",
       "6            0  \n",
       "7            0  \n",
       "8            0  \n",
       "9            0  \n",
       "10           0  \n",
       "11           0  \n",
       "12           0  \n",
       "13           0  \n",
       "14           0  \n",
       "15           0  \n",
       "16           0  \n",
       "17           0  \n",
       "18           0  \n",
       "19           0  \n",
       "20           0  \n",
       "21           0  \n",
       "22           0  \n",
       "23           0  \n",
       "24           0  \n",
       "25           0  \n",
       "26           0  \n",
       "27           0  \n",
       "28           0  \n",
       "29           0  \n",
       "...        ...  \n",
       "149970       0  \n",
       "149971       0  \n",
       "149972       0  \n",
       "149973       0  \n",
       "149974       0  \n",
       "149975       0  \n",
       "149976       0  \n",
       "149977       0  \n",
       "149978       0  \n",
       "149979       0  \n",
       "149980       0  \n",
       "149981       0  \n",
       "149982       0  \n",
       "149983       0  \n",
       "149984       0  \n",
       "149985       0  \n",
       "149986       0  \n",
       "149987       0  \n",
       "149988       0  \n",
       "149989       0  \n",
       "149990       0  \n",
       "149991       0  \n",
       "149992       0  \n",
       "149993       0  \n",
       "149994       0  \n",
       "149995       0  \n",
       "149996       0  \n",
       "149997       0  \n",
       "149998       0  \n",
       "149999       0  \n",
       "\n",
       "[150000 rows x 28 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_scaled = df.copy()\n",
    "scaler = StandardScaler()\n",
    "\n",
    "df_scaled[features] = pd.DataFrame(scaler.fit_transform(df_scaled[features]), columns=features)\n",
    "\n",
    "df_scaled"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### PCA Explained Variance Cumsum"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pca = PCA()\n",
    "pca.fit(df_scaled)\n",
    "cumsum = np.cumsum(pca.explained_variance_ratio_)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(12,8))\n",
    "\n",
    "ax.set_title('Explained Variance x Number of Dimensions')\n",
    "ax.set_xlabel('Dimensions')\n",
    "ax.set_ylabel('Explained Variance')\n",
    "plt.plot(range(len(cumsum)), cumsum)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PCA is not able to retain much variance with few features (above 80% with 20 dimensions), so, it is not useful for this dataset and will not be used. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling <a class=\"anchor\" id=\"8\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For modeling, I will adopt a stratified 70/30 split and compare Random Forest, Naive Bayes, Logistic Regression and SVM on the Standardized dataset (required for SVM and LR, not necessary for RF and NB). Since it is a Fraud detection problem, the focus will be on **Recall**, since it is much worse to not detect a fraudster than to have a false alarm on non-fraudsters (*precision*)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Splitting Test/Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fraud distribution in Training Set: Counter({0: 104834, 1: 166}), 0.16%\n",
      "Fraud distribution in Test Set: Counter({0: 44929, 1: 71}), 0.16%\n"
     ]
    }
   ],
   "source": [
    "X = df_scaled[features].values \n",
    "y = df_scaled['Fraude'].values\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42, stratify=y)\n",
    "\n",
    "print('Fraud distribution in Training Set: {}, {:.2%}'.format(Counter(y_train), \n",
    "                                                      Counter(y_train)[1]/sum(Counter(y_train).values())))\n",
    "print('Fraud distribution in Test Set: {}, {:.2%}'.format(Counter(y_test), \n",
    "                                                      Counter(y_test)[1]/sum(Counter(y_test).values())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train RandomForest "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest trained!\n",
      "\n"
     ]
    }
   ],
   "source": [
    "rf = RandomForestClassifier(n_estimators=10, random_state=42)  \n",
    "rf.fit(X_train, y_train)  \n",
    "\n",
    "y_pred_train = rf.predict(X_train)\n",
    "y_pred_test = rf.predict(X_test)\n",
    "\n",
    "print(\"Random Forest trained!\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Simple Metrics Function (CM, Acc, AUC and KS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def model_metrics (y_actual, y_predicted):\n",
    "    #CM\n",
    "    print(\"Confusion Matrix: \\n\", confusion_matrix(y_actual, y_predicted))  \n",
    "    print(\"Classification Report: \\n\", classification_report(y_actual, y_predicted))  \n",
    "    print(\"Accuracy: \", accuracy_score(y_actual, y_predicted))   \n",
    "\n",
    "    #AUC and KS\n",
    "    print(\"AUC: \", roc_auc_score(y_actual, y_predicted))\n",
    "    print(\"KS: \", ks_2samp(y_predicted[y_actual == 0], y_predicted[y_actual == 1]).statistic)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Performance on Training Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[104833      1]\n",
      " [    10    156]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00    104834\n",
      "           1       0.99      0.94      0.97       166\n",
      "\n",
      "   micro avg       1.00      1.00      1.00    105000\n",
      "   macro avg       1.00      0.97      0.98    105000\n",
      "weighted avg       1.00      1.00      1.00    105000\n",
      "\n",
      "Accuracy:  0.9998952380952381\n",
      "AUC:  0.9698747486272619\n",
      "KS:  0.9397494972545236\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Performance on Test Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[44925     4]\n",
      " [   23    48]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     44929\n",
      "           1       0.92      0.68      0.78        71\n",
      "\n",
      "   micro avg       1.00      1.00      1.00     45000\n",
      "   macro avg       0.96      0.84      0.89     45000\n",
      "weighted avg       1.00      1.00      1.00     45000\n",
      "\n",
      "Accuracy:  0.9994\n",
      "AUC:  0.8379836543353693\n",
      "KS:  0.6759673086707384\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_test, y_pred_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Random Forest does really well on the training set, but suffers from great overfitting. Since it is an imbalanced problem, we can't look at accuracy, but rather KS and Recall. Those metric are only 0.68, although precision is good (rendering a respectable 0.78 F1-Score). If one wanted to maximize F1, this model is not too shabby, but we are looking for good Recall *regardless* of precision."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Feature Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "feature_importances = pd.DataFrame(rf.feature_importances_,\n",
    "                                index = features,\n",
    "                                    columns=['importance']).sort_values('importance', ascending=False)\n",
    "\n",
    "plt.figure(figsize=(12,6))\n",
    "feature_importances.importance[:10].plot(kind='barh', color='green')\n",
    "plt.title('Feature Importance - Top 10')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that 4 features are predominant (PP17, PP12, PP14 and PP11). Not coincidentally, those features are among the ones with the highest correlation to the target (review correlations heatmap if necessary). The model makes sense."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "importance    0.605623\n",
       "dtype: float64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_importances.iloc[0:4].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Those 4 variables are responsible for 60% of the splits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PP8</th>\n",
       "      <td>0.010476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP20</th>\n",
       "      <td>0.010262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP15</th>\n",
       "      <td>0.009730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sacado</th>\n",
       "      <td>0.009017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PP24</th>\n",
       "      <td>0.002828</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        importance\n",
       "PP8       0.010476\n",
       "PP20      0.010262\n",
       "PP15      0.009730\n",
       "Sacado    0.009017\n",
       "PP24      0.002828"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_importances.iloc[-5:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PP24 has the lowest importance, it's the third variable with the highest variance, has 0 correlation with the target and 0 fraud frequency in its outliers. Hence, it is most probably just noise confusing our algorithms. Will drop this column. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Drop PP24 and check if RF does slightly better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[44925     4]\n",
      " [   24    47]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     44929\n",
      "           1       0.92      0.66      0.77        71\n",
      "\n",
      "   micro avg       1.00      1.00      1.00     45000\n",
      "   macro avg       0.96      0.83      0.89     45000\n",
      "weighted avg       1.00      1.00      1.00     45000\n",
      "\n",
      "Accuracy:  0.9993777777777778\n",
      "AUC:  0.8309414008142425\n",
      "KS:  0.6618828016284849\n"
     ]
    }
   ],
   "source": [
    "df_new = df_scaled.copy()\n",
    "\n",
    "df_new.drop(['PP24'], axis=1, inplace=True)\n",
    "features = [col for col in df_new if col != 'Fraude']\n",
    "\n",
    "X = df_new[features].values \n",
    "y = df_new['Fraude'].values\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42, stratify=y)\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=10, random_state=42)  \n",
    "rf.fit(X_train, y_train)  \n",
    "\n",
    "y_pred_test = rf.predict(X_test)\n",
    "\n",
    "model_metrics(y_test, y_pred_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, it just made one more fraudster being incorrectly classified as clean. Small change, but for the worst. Will keep variable PP24."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Try Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gaussian NB trained!\n",
      "\n"
     ]
    }
   ],
   "source": [
    "nb = GaussianNB() \n",
    "nb.fit(X_train, y_train)  \n",
    "\n",
    "y_pred_train = nb.predict(X_train)\n",
    "y_pred_test = nb.predict(X_test)\n",
    "\n",
    "print(\"Gaussian NB trained!\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[102971   1863]\n",
      " [    33    133]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.98      0.99    104834\n",
      "           1       0.07      0.80      0.12       166\n",
      "\n",
      "   micro avg       0.98      0.98      0.98    105000\n",
      "   macro avg       0.53      0.89      0.56    105000\n",
      "weighted avg       1.00      0.98      0.99    105000\n",
      "\n",
      "Accuracy:  0.9819428571428571\n",
      "AUC:  0.8917169335525515\n",
      "KS:  0.7834338671051031\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[44103   826]\n",
      " [   11    60]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.98      0.99     44929\n",
      "           1       0.07      0.85      0.13        71\n",
      "\n",
      "   micro avg       0.98      0.98      0.98     45000\n",
      "   macro avg       0.53      0.91      0.56     45000\n",
      "weighted avg       1.00      0.98      0.99     45000\n",
      "\n",
      "Accuracy:  0.9814\n",
      "AUC:  0.9133429301128948\n",
      "KS:  0.8266858602257897\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_test, y_pred_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although NB does not perform as well as RF on the training set, it generalizes much better, and has a much higher Recall (0.85!). The \"cost\", however, is a huge decrease on Precision (0.07) and consequently on F-1 (0.13). Again, this depends on the objective - since we are focusing on Recall no matter what, NB did better than RF."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Try Regularized Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR trained!\n",
      "\n"
     ]
    }
   ],
   "source": [
    "lr = LogisticRegression(penalty='l2', C=1.0, solver='lbfgs', random_state=42) \n",
    "lr.fit(X_train, y_train)  \n",
    "\n",
    "y_pred_train = lr.predict(X_train)\n",
    "y_pred_test = lr.predict(X_test)\n",
    "\n",
    "print(\"LR trained!\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[104824     10]\n",
      " [    41    125]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00    104834\n",
      "           1       0.93      0.75      0.83       166\n",
      "\n",
      "   micro avg       1.00      1.00      1.00    105000\n",
      "   macro avg       0.96      0.88      0.92    105000\n",
      "weighted avg       1.00      1.00      1.00    105000\n",
      "\n",
      "Accuracy:  0.9995142857142857\n",
      "AUC:  0.8764583296461118\n",
      "KS:  0.7529166592922236\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[44926     3]\n",
      " [   23    48]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     44929\n",
      "           1       0.94      0.68      0.79        71\n",
      "\n",
      "   micro avg       1.00      1.00      1.00     45000\n",
      "   macro avg       0.97      0.84      0.89     45000\n",
      "weighted avg       1.00      1.00      1.00     45000\n",
      "\n",
      "Accuracy:  0.9994222222222222\n",
      "AUC:  0.837994783005048\n",
      "KS:  0.6759895660100961\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_test, y_pred_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic Regression generalized better than RF, but had similar results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Try Linear SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Linear SVM trained!\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "lsvm = LinearSVC(penalty='l2', loss='hinge', C=1.0, max_iter=5000, random_state=42) \n",
    "lsvm.fit(X_train, y_train)  \n",
    "\n",
    "y_pred_train = lsvm.predict(X_train)\n",
    "y_pred_test = lsvm.predict(X_test)\n",
    "\n",
    "print(\"Linear SVM trained!\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[104815     19]\n",
      " [    39    127]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00    104834\n",
      "           1       0.87      0.77      0.81       166\n",
      "\n",
      "   micro avg       1.00      1.00      1.00    105000\n",
      "   macro avg       0.93      0.88      0.91    105000\n",
      "weighted avg       1.00      1.00      1.00    105000\n",
      "\n",
      "Accuracy:  0.999447619047619\n",
      "AUC:  0.8824395010264076\n",
      "KS:  0.7648790020528151\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_train, y_pred_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion Matrix: \n",
      " [[44919    10]\n",
      " [   20    51]]\n",
      "Classification Report: \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00     44929\n",
      "           1       0.84      0.72      0.77        71\n",
      "\n",
      "   micro avg       1.00      1.00      1.00     45000\n",
      "   macro avg       0.92      0.86      0.89     45000\n",
      "weighted avg       1.00      1.00      1.00     45000\n",
      "\n",
      "Accuracy:  0.9993333333333333\n",
      "AUC:  0.8590436428806765\n",
      "KS:  0.718087285761353\n"
     ]
    }
   ],
   "source": [
    "model_metrics(y_test, y_pred_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SVM also generalized better than the Random Forest and had basically the same F1-Score (0.77) - with a better Recall, AUC and KS, and worse Precision. If one wanted F-1 as metric, SVM would be the choice."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### GridSearch CV <a class=\"anchor\" id=\"9\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lastly, I will try a GridSearch using 5-fold Cross Validation and optimizing Recall - to see if Naive Bayes can be beaten."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 301,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.735\n",
      "Best parameters set:\n",
      "\tcriterion: 'entropy'\n",
      "\tmax_depth: 10\n",
      "\tmin_samples_leaf: 5\n",
      "\tn_estimators: 100\n"
     ]
    }
   ],
   "source": [
    "rf_parameters = {\n",
    "    'n_estimators': [10, 50, 100],\n",
    "    'criterion': ['gini', 'entropy'],\n",
    "    'max_depth': [2, 5, 10, 50, 200],\n",
    "    'min_samples_leaf': [5, 10, 50]\n",
    "    }\n",
    "\n",
    "gsrf = GridSearchCV(RandomForestClassifier(random_state=42), param_grid=rf_parameters, cv=5, \n",
    "                    scoring='recall')\n",
    "\n",
    "gsrf.fit(X_train, y_train)\n",
    "\n",
    "print(\"Best score: %0.3f\" % gsrf.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = gsrf.best_estimator_.get_params()\n",
    "for param_name in sorted(rf_parameters.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "About 7 points of increase, but still far from NB."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.729\n",
      "Best parameters set:\n",
      "\tC: 100\n",
      "\tpenalty: 'l1'\n"
     ]
    }
   ],
   "source": [
    "lr_parameters = {\n",
    "    'penalty': ['l1', 'l2'],\n",
    "    'C': [100, 10, 1, 0.1, 0.01]\n",
    "    }\n",
    "\n",
    "gslr = GridSearchCV(LogisticRegression(random_state=42), param_grid=lr_parameters, cv=5, \n",
    "                    scoring='recall')\n",
    "\n",
    "gslr.fit(X_train, y_train)\n",
    "\n",
    "print(\"Best score: %0.3f\" % gslr.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = gslr.best_estimator_.get_params()\n",
    "for param_name in sorted(lr_parameters.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once again, very similar to Random Forest."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RBF SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.717\n",
      "Best parameters set:\n",
      "\tC: 1\n",
      "\tgamma: 0.01\n"
     ]
    }
   ],
   "source": [
    "svm_parameters = {\n",
    "    'gamma': [1, 0.1, 0.01],\n",
    "    'C': [100, 10, 1, 0.01]\n",
    "    }\n",
    "\n",
    "gssvm = GridSearchCV(SVC(random_state=42, kernel='rbf'), param_grid=svm_parameters, cv=5, \n",
    "                    scoring='recall')\n",
    "\n",
    "gssvm.fit(X_train, y_train)\n",
    "\n",
    "print(\"Best score: %0.3f\" % gssvm.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = gssvm.best_estimator_.get_params()\n",
    "for param_name in sorted(svm_parameters.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Linear SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n",
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.753\n",
      "Best parameters set:\n",
      "\tC: 100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/sklearn/svm/base.py:922: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.\n",
      "  \"the number of iterations.\", ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "svm_parameters = {\n",
    "    'C': [1000, 100, 10, 1, 0.01]\n",
    "    }\n",
    "\n",
    "gssvm = GridSearchCV(LinearSVC(random_state=42), param_grid=svm_parameters, cv=2, \n",
    "                    scoring='recall')\n",
    "\n",
    "gssvm.fit(X_train, y_train)\n",
    "\n",
    "print(\"Best score: %0.3f\" % gssvm.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = gssvm.best_estimator_.get_params()\n",
    "for param_name in sorted(svm_parameters.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion <a class=\"anchor\" id=\"10\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We started with a dataset of 150k records and 30 features, containing an unbalanced (0.16%) amount of fraudsters and no null values. Most of the features were normally distributed and had very little correlation. Outliers (|Z-score| > 3) with no fraudster frequency were imputed with the mean, variables with low variance (PCA was also tried but deemed not useful, as too little variance was kept in low dimension spaces) were removed and then scaled.\n",
    "\n",
    "For modeling, a stratified 70/30 split was adopted and Recall was the most important metric observed. Out of the models tested (Random Forest, Naive Bayes, Logistic Regression and SVM), NB had the best performance, achieving 0.85 Recall while not overfitting (although its precision and F1-score were much lower than the other algorithms - but we are focusing on Recall solely). While RF was almost perfect on the training set, it severely overfitted and had a Recall of 0.68 (same as LR, that generalized better). Linear SVM had about the same F-1 Score (0.77) as LR and RF, but it has a higher Recall (0.72) and Lower Precision (0.84) - hence, would be prefered if F-1 was the metric chosen. GridSearch did improve some points in Recall, but still about 10 points below NB.\n",
    "\n",
    "If one wanted to go even further on this analysis, techniques such as Undersampling, Oversampling and Synthetic Sampling (like SMOTE) could be used to avert the imbalance between classes. Even more, the problem could be turned into an Anomaly Detection, using techniques such as Isolation Forest or Local Outlier Factor."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}