Effect of Earnings Reports on Stock Price.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import plotly.graph_objects as go\n",
    "import plotly\n",
    "from scipy.stats import ttest_ind"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/stock_filing_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>filing_date</th>\n",
       "      <th>gics_sector</th>\n",
       "      <th>company_id</th>\n",
       "      <th>date</th>\n",
       "      <th>filing_released</th>\n",
       "      <th>n_calendar</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>close</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>percent_growth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-29</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>198.2300</td>\n",
       "      <td>0.001971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-26</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>201.3000</td>\n",
       "      <td>0.017489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-23</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>200.0900</td>\n",
       "      <td>0.011373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-24</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>197.7800</td>\n",
       "      <td>-0.000303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-25</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>197.8400</td>\n",
       "      <td>197.5100</td>\n",
       "      <td>-0.001668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>15237</td>\n",
       "      <td>2018-01-29</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>254.2695</td>\n",
       "      <td>249.0841</td>\n",
       "      <td>-0.020393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>29618</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>156.5710</td>\n",
       "      <td>156.5710</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>15237</td>\n",
       "      <td>2018-01-26</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>254.2695</td>\n",
       "      <td>254.2695</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>15237</td>\n",
       "      <td>2018-01-25</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>254.2695</td>\n",
       "      <td>245.5300</td>\n",
       "      <td>-0.034371</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  filing_date             gics_sector  company_id        date  \\\n",
       "0  2018-01-22  Information Technology        2236  2018-01-29   \n",
       "1  2018-01-22  Information Technology        2236  2018-01-26   \n",
       "2  2018-01-22  Information Technology        2236  2018-01-23   \n",
       "3  2018-01-22  Information Technology        2236  2018-01-22   \n",
       "4  2018-01-22  Information Technology        2236  2018-01-24   \n",
       "5  2018-01-22  Information Technology        2236  2018-01-25   \n",
       "6  2018-01-22  Information Technology       15237  2018-01-29   \n",
       "7  2018-01-22  Information Technology       29618  2018-01-22   \n",
       "8  2018-01-22  Information Technology       15237  2018-01-26   \n",
       "9  2018-01-22  Information Technology       15237  2018-01-25   \n",
       "\n",
       "   filing_released  n_calendar  n_trading     close  close_next_n  \\\n",
       "0                1           7          5  197.8400      198.2300   \n",
       "1                1           4          4  197.8400      201.3000   \n",
       "2                1           1          1  197.8400      200.0900   \n",
       "3                1           0          0  197.8400      197.8400   \n",
       "4                1           2          2  197.8400      197.7800   \n",
       "5                1           3          3  197.8400      197.5100   \n",
       "6                0           7          5  254.2695      249.0841   \n",
       "7                0           0          0  156.5710      156.5710   \n",
       "8                0           4          4  254.2695      254.2695   \n",
       "9                0           3          3  254.2695      245.5300   \n",
       "\n",
       "   percent_growth  \n",
       "0        0.001971  \n",
       "1        0.017489  \n",
       "2        0.011373  \n",
       "3        0.000000  \n",
       "4       -0.000303  \n",
       "5       -0.001668  \n",
       "6       -0.020393  \n",
       "7        0.000000  \n",
       "8        0.000000  \n",
       "9       -0.034371  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>company_id</th>\n",
       "      <th>filing_released</th>\n",
       "      <th>n_calendar</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>close</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>percent_growth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>count</td>\n",
       "      <td>145880.000000</td>\n",
       "      <td>145880.000000</td>\n",
       "      <td>145880.000000</td>\n",
       "      <td>145880.000000</td>\n",
       "      <td>145880.000000</td>\n",
       "      <td>145880.000000</td>\n",
       "      <td>145880.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>mean</td>\n",
       "      <td>9176.696031</td>\n",
       "      <td>0.037565</td>\n",
       "      <td>3.554833</td>\n",
       "      <td>2.401529</td>\n",
       "      <td>116.304532</td>\n",
       "      <td>116.431583</td>\n",
       "      <td>0.000980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>std</td>\n",
       "      <td>10176.751878</td>\n",
       "      <td>0.190143</td>\n",
       "      <td>2.516643</td>\n",
       "      <td>1.666983</td>\n",
       "      <td>159.411843</td>\n",
       "      <td>159.828379</td>\n",
       "      <td>0.028021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>min</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.758800</td>\n",
       "      <td>6.637100</td>\n",
       "      <td>-0.384850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25%</td>\n",
       "      <td>1306.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>48.955550</td>\n",
       "      <td>49.008025</td>\n",
       "      <td>-0.008891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50%</td>\n",
       "      <td>4997.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>82.094900</td>\n",
       "      <td>82.173400</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75%</td>\n",
       "      <td>14221.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>136.558300</td>\n",
       "      <td>136.465000</td>\n",
       "      <td>0.012031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>max</td>\n",
       "      <td>35080.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>2188.900000</td>\n",
       "      <td>2188.900000</td>\n",
       "      <td>0.582145</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          company_id  filing_released     n_calendar      n_trading  \\\n",
       "count  145880.000000    145880.000000  145880.000000  145880.000000   \n",
       "mean     9176.696031         0.037565       3.554833       2.401529   \n",
       "std     10176.751878         0.190143       2.516643       1.666983   \n",
       "min         6.000000         0.000000       0.000000       0.000000   \n",
       "25%      1306.000000         0.000000       1.000000       1.000000   \n",
       "50%      4997.000000         0.000000       4.000000       2.000000   \n",
       "75%     14221.000000         0.000000       6.000000       4.000000   \n",
       "max     35080.000000         1.000000       7.000000       5.000000   \n",
       "\n",
       "               close   close_next_n  percent_growth  \n",
       "count  145880.000000  145880.000000   145880.000000  \n",
       "mean      116.304532     116.431583        0.000980  \n",
       "std       159.411843     159.828379        0.028021  \n",
       "min         6.758800       6.637100       -0.384850  \n",
       "25%        48.955550      49.008025       -0.008891  \n",
       "50%        82.094900      82.173400        0.000000  \n",
       "75%       136.558300     136.465000        0.012031  \n",
       "max      2188.900000    2188.900000        0.582145  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(df.head(10))\n",
    "display(df.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset Overview\n",
    "\n",
    "* Samples: 145,880\n",
    "* Timeline\n",
    "    * From 2018-01-01\n",
    "    * To 2019-12-27\n",
    "* Number of Industry Sectors: 11\n",
    "* Companies: 479\n",
    "* Sampling Pool: S&P 500\n",
    "\n",
    "### Features\n",
    "\n",
    "* filing_date: Date when annual report (10-k) has been published\n",
    "* gics_sector: Sector (Industry) where the company primarily conducts business\n",
    "* company_id: Company Identifier (in this study we don't disclose company names)\n",
    "* date: Calendar Date (always ahead of filing_date) at which 'close_next_n' has been recorded\n",
    "* filing_released: Boolean flag indicating whther a company has released a report as of the row's filing_date\n",
    "* n_calendar: Number of calendar days since filing_date\n",
    "* n_trading: Number of trading days (weekdays excluding holidays) since filing_date\n",
    "* close: Closing price as of 'filing_date'\n",
    "* close_next_n: Closing price as of 'date'\n",
    "* percent_growth: Percent change (growth) in price from 'close' to 'close_next_n'\n",
    "\n",
    "\n",
    "### How it was built\n",
    "\n",
    "This dataset was generated to study the effect of financial statement release (10-K report) on the behaviour of stock prices of the corresponding company shortly after the report became public. Population of this dataset started with identifying companies that released their financial report and date of their filings, generating unique combinations of company_id, filing_date and filing_released=1. For each such combination, complimentary combinations of companies that did not release their financial report (filing_released=0) at filinng_date but belonged to the same industry were generated. Finally, each combination was extended (duplicated) to cover the timeline of date = [filing_date, filing_date + 7 calendar days] for each company, keeping track of stock prices at those dates.\n",
    "\n",
    "### Lets get started!\n",
    "We want to see price change between sample at n=1 and sample n=1+shift (several days ahead of n=1). To do this, we will create df_shift and update 'n' column to become 'n' minus 'shift'. This way, we can join df_ and df_shift so that df_[n] is in the same row with df[n+shift]."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>filing_date</th>\n",
       "      <th>gics_sector</th>\n",
       "      <th>company_id</th>\n",
       "      <th>date</th>\n",
       "      <th>filing_released</th>\n",
       "      <th>n_calendar</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>close</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>percent_growth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-29</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>197.84</td>\n",
       "      <td>198.23</td>\n",
       "      <td>0.001971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-26</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>197.84</td>\n",
       "      <td>201.30</td>\n",
       "      <td>0.017489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-23</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>197.84</td>\n",
       "      <td>200.09</td>\n",
       "      <td>0.011373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>197.84</td>\n",
       "      <td>197.84</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-24</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>197.84</td>\n",
       "      <td>197.78</td>\n",
       "      <td>-0.000303</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  filing_date             gics_sector  company_id        date  \\\n",
       "0  2018-01-22  Information Technology        2236  2018-01-29   \n",
       "1  2018-01-22  Information Technology        2236  2018-01-26   \n",
       "2  2018-01-22  Information Technology        2236  2018-01-23   \n",
       "3  2018-01-22  Information Technology        2236  2018-01-22   \n",
       "4  2018-01-22  Information Technology        2236  2018-01-24   \n",
       "\n",
       "   filing_released  n_calendar  n_trading   close  close_next_n  \\\n",
       "0                1           7          5  197.84        198.23   \n",
       "1                1           4          4  197.84        201.30   \n",
       "2                1           1          1  197.84        200.09   \n",
       "3                1           0          0  197.84        197.84   \n",
       "4                1           2          2  197.84        197.78   \n",
       "\n",
       "   percent_growth  \n",
       "0        0.001971  \n",
       "1        0.017489  \n",
       "2        0.011373  \n",
       "3        0.000000  \n",
       "4       -0.000303  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>filing_date</th>\n",
       "      <th>gics_sector</th>\n",
       "      <th>company_id</th>\n",
       "      <th>date</th>\n",
       "      <th>filing_released</th>\n",
       "      <th>n_calendar</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>close</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>percent_growth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-29</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>197.84</td>\n",
       "      <td>198.23</td>\n",
       "      <td>0.001971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-26</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>197.84</td>\n",
       "      <td>201.30</td>\n",
       "      <td>0.017489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-23</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>197.84</td>\n",
       "      <td>200.09</td>\n",
       "      <td>0.011373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-2</td>\n",
       "      <td>197.84</td>\n",
       "      <td>197.84</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>2018-01-22</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2236</td>\n",
       "      <td>2018-01-24</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>197.84</td>\n",
       "      <td>197.78</td>\n",
       "      <td>-0.000303</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  filing_date             gics_sector  company_id        date  \\\n",
       "0  2018-01-22  Information Technology        2236  2018-01-29   \n",
       "1  2018-01-22  Information Technology        2236  2018-01-26   \n",
       "2  2018-01-22  Information Technology        2236  2018-01-23   \n",
       "3  2018-01-22  Information Technology        2236  2018-01-22   \n",
       "4  2018-01-22  Information Technology        2236  2018-01-24   \n",
       "\n",
       "   filing_released  n_calendar  n_trading   close  close_next_n  \\\n",
       "0                1           7          3  197.84        198.23   \n",
       "1                1           4          2  197.84        201.30   \n",
       "2                1           1         -1  197.84        200.09   \n",
       "3                1           0         -2  197.84        197.84   \n",
       "4                1           2          0  197.84        197.78   \n",
       "\n",
       "   percent_growth  \n",
       "0        0.001971  \n",
       "1        0.017489  \n",
       "2        0.011373  \n",
       "3        0.000000  \n",
       "4       -0.000303  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shift = 2\n",
    "df_ = df.copy()\n",
    "df_shift = df_.copy()\n",
    "df_shift['n_trading'] = df_shift['n_trading'] - shift\n",
    "display(df_.head())\n",
    "display(df_shift.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Note</b> above how entries with the same index have different values in the 'n' column.\n",
    "In the following step, we will join df_ and df_shift using 'company_id', 'filing_date' and 'n_trading'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size before merge: 145880\n",
      "Size after merge: 95312\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>company_id</th>\n",
       "      <th>filing_date</th>\n",
       "      <th>date</th>\n",
       "      <th>date_after_shift_days</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>close_next_n_after_shift_days</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>15071</td>\n",
       "      <td>2019-11-18</td>\n",
       "      <td>2019-11-19</td>\n",
       "      <td>2019-11-21</td>\n",
       "      <td>99.0381</td>\n",
       "      <td>97.7299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>11050</td>\n",
       "      <td>2018-02-23</td>\n",
       "      <td>2018-02-28</td>\n",
       "      <td>2018-03-02</td>\n",
       "      <td>166.0300</td>\n",
       "      <td>170.0600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6729</td>\n",
       "      <td>2019-02-25</td>\n",
       "      <td>2019-02-25</td>\n",
       "      <td>2019-02-27</td>\n",
       "      <td>39.9900</td>\n",
       "      <td>39.9300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>1619</td>\n",
       "      <td>2018-01-25</td>\n",
       "      <td>2018-01-29</td>\n",
       "      <td>2018-01-31</td>\n",
       "      <td>40.4542</td>\n",
       "      <td>40.3040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>726</td>\n",
       "      <td>2019-01-25</td>\n",
       "      <td>2019-01-30</td>\n",
       "      <td>2019-02-01</td>\n",
       "      <td>105.9312</td>\n",
       "      <td>106.7729</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   company_id filing_date        date date_after_shift_days  close_next_n  \\\n",
       "0       15071  2019-11-18  2019-11-19            2019-11-21       99.0381   \n",
       "1       11050  2018-02-23  2018-02-28            2018-03-02      166.0300   \n",
       "2        6729  2019-02-25  2019-02-25            2019-02-27       39.9900   \n",
       "3        1619  2018-01-25  2018-01-29            2018-01-31       40.4542   \n",
       "4         726  2019-01-25  2019-01-30            2019-02-01      105.9312   \n",
       "\n",
       "   close_next_n_after_shift_days  \n",
       "0                        97.7299  \n",
       "1                       170.0600  \n",
       "2                        39.9300  \n",
       "3                        40.3040  \n",
       "4                       106.7729  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index_columns = ['company_id', 'filing_date', 'n_trading']\n",
    "print(f'Size before merge: {len(df_)}')\n",
    "df_merged = df_.set_index(index_columns).join(\n",
    "    df_shift.set_index(index_columns), \n",
    "    how='inner',\n",
    "    rsuffix='_after_shift_days'\n",
    ").reset_index()\n",
    "print(f'Size after merge: {len(df_merged)}')\n",
    "df_merged[['company_id', 'filing_date', 'date', 'date_after_shift_days', 'close_next_n', 'close_next_n_after_shift_days']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>To make sure</b> I've done the join correctly, we can see above how 'date' is 2 days before 'date_after_shift_days'. The size the dataframe shrinks from 145,880 to 95,312 entries due to the fact that n_trading differs between df_merge [0:5] and df_shift [-2:3].\n",
    "\n",
    "<b>Next</b>, we're going to filter the table to n (n_trading) = 1, leaving ourselves with entries corresponding to exactly 1 day since a financial statement report has been released. Rows where n=1 will also have information on n=3, coming from the above join. We will also filter out entries where price growth during the first day was less than 1%, assuming there was weak momentum and the public did not feel particularly good about the filing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>company_id</th>\n",
       "      <th>filing_date</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>percent_growth</th>\n",
       "      <th>price_growth_during_shift_days</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>15071</td>\n",
       "      <td>2019-11-18</td>\n",
       "      <td>1</td>\n",
       "      <td>0.018694</td>\n",
       "      <td>-0.013209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>2203</td>\n",
       "      <td>2019-02-08</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.008674</td>\n",
       "      <td>0.006226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>477</td>\n",
       "      <td>2019-02-26</td>\n",
       "      <td>1</td>\n",
       "      <td>0.001636</td>\n",
       "      <td>0.003675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>31206</td>\n",
       "      <td>2018-02-01</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.007800</td>\n",
       "      <td>0.188938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>7345</td>\n",
       "      <td>2018-02-09</td>\n",
       "      <td>1</td>\n",
       "      <td>0.034449</td>\n",
       "      <td>0.023006</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    company_id filing_date  n_trading  percent_growth  \\\n",
       "0        15071  2019-11-18          1        0.018694   \n",
       "8         2203  2019-02-08          1       -0.008674   \n",
       "16         477  2019-02-26          1        0.001636   \n",
       "17       31206  2018-02-01          1       -0.007800   \n",
       "19        7345  2018-02-09          1        0.034449   \n",
       "\n",
       "    price_growth_during_shift_days  \n",
       "0                        -0.013209  \n",
       "8                         0.006226  \n",
       "16                        0.003675  \n",
       "17                        0.188938  \n",
       "19                        0.023006  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_merged = df_merged[df_merged['n_trading'] == 1]\n",
    "df_merged['growth_increment'] = df_merged['percent_growth_after_shift_days'] - df_merged['percent_growth']\n",
    "\n",
    "# Computing how price changed at n=3 compared to n=1 (percent change)\n",
    "df_merged['price_growth_during_shift_days'] = df_merged['close_next_n_after_shift_days'] / df_merged['close_next_n'] - 1\n",
    "display(df_merged[['company_id', 'filing_date', 'n_trading', 'percent_growth', 'price_growth_during_shift_days']].head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Note</b>: 'percent_growth' represents how price grew between n=0 and n=1 (initial reaction to the financial statement release) while 'price_growth_during_shift_days' represents price change between n=1 and n=3.\n",
    "\n",
    "This completes our data wrangling phase. At this point, we can compare whether the event of financial statement release can be used to say anything about how price will behave in the near future, in this case between day 1 and day 3 after the statement became public.\n",
    "\n",
    "We will also apply a threshold of 0.01 (1% stock price increase) on 'percent_growth' for companies that released their filings to identify companies that had strong momentum in the first day of the statement release."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's define a printing function (don't focus too much on this one - it just makes a nice histogram)\n",
    "def print_hist(positive, negative, title=''):\n",
    "    \"\"\" Prints an overlapping histogram of two arrays \"\"\"\n",
    "\n",
    "    fig = go.Figure()\n",
    "    \n",
    "    fig.add_trace(go.Histogram(\n",
    "        x=positive,\n",
    "        histnorm='percent',\n",
    "        name='Filing Released', # name used in legend and hover labels\n",
    "        xbins=dict( # bins used for histogram\n",
    "            start=-0.1,\n",
    "            end=0.1,\n",
    "            size=0.01\n",
    "        ),\n",
    "        marker_color='#EB89B5',\n",
    "        opacity=0.75\n",
    "    ))\n",
    "    \n",
    "    fig.add_trace(go.Histogram(\n",
    "        x=negative,\n",
    "        histnorm='percent',\n",
    "        name='Ordinary Day',\n",
    "        xbins=dict(\n",
    "            start=-0.1,\n",
    "            end=0.1,\n",
    "            size=0.01\n",
    "        ),\n",
    "        marker_color='#330C73',\n",
    "        opacity=0.75\n",
    "    ))\n",
    "    \n",
    "    fig.update_layout(\n",
    "        title_text=title, # title of plot\n",
    "        xaxis_title_text='Price Growth during [2, t] days since the Fianancial Statement Release', # xaxis label\n",
    "        yaxis_title_text='Samples, %', # yaxis label\n",
    "        bargap=0.2, # gap between bars of adjacent location coordinates\n",
    "        bargroupgap=0.1 # gap between bars of the same location coordinates\n",
    "    )\n",
    "    \n",
    "    fig.show(\"png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Note:</b> Logistics out of the way. Now to the actual code!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>company_id</th>\n",
       "      <th>filing_date</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>gics_sector</th>\n",
       "      <th>date</th>\n",
       "      <th>filing_released</th>\n",
       "      <th>n_calendar</th>\n",
       "      <th>close</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>percent_growth</th>\n",
       "      <th>gics_sector_after_shift_days</th>\n",
       "      <th>date_after_shift_days</th>\n",
       "      <th>filing_released_after_shift_days</th>\n",
       "      <th>n_calendar_after_shift_days</th>\n",
       "      <th>close_after_shift_days</th>\n",
       "      <th>close_next_n_after_shift_days</th>\n",
       "      <th>percent_growth_after_shift_days</th>\n",
       "      <th>growth_increment</th>\n",
       "      <th>price_growth_during_shift_days</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>62</td>\n",
       "      <td>31816</td>\n",
       "      <td>2019-02-14</td>\n",
       "      <td>1</td>\n",
       "      <td>Health Care</td>\n",
       "      <td>2019-02-15</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>92.8511</td>\n",
       "      <td>94.9518</td>\n",
       "      <td>0.022624</td>\n",
       "      <td>Health Care</td>\n",
       "      <td>2019-02-20</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>92.8511</td>\n",
       "      <td>93.9761</td>\n",
       "      <td>0.012116</td>\n",
       "      <td>-0.010508</td>\n",
       "      <td>-0.010276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>401</td>\n",
       "      <td>1070</td>\n",
       "      <td>2019-02-15</td>\n",
       "      <td>1</td>\n",
       "      <td>Materials</td>\n",
       "      <td>2019-02-19</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>12.1166</td>\n",
       "      <td>12.9159</td>\n",
       "      <td>0.065967</td>\n",
       "      <td>Materials</td>\n",
       "      <td>2019-02-21</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>12.1166</td>\n",
       "      <td>12.8468</td>\n",
       "      <td>0.060264</td>\n",
       "      <td>-0.005703</td>\n",
       "      <td>-0.005350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1166</td>\n",
       "      <td>29618</td>\n",
       "      <td>2018-10-24</td>\n",
       "      <td>1</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2018-10-25</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>151.2189</td>\n",
       "      <td>154.9717</td>\n",
       "      <td>0.024817</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2018-10-29</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>151.2189</td>\n",
       "      <td>150.8436</td>\n",
       "      <td>-0.002482</td>\n",
       "      <td>-0.027299</td>\n",
       "      <td>-0.026638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1450</td>\n",
       "      <td>5060</td>\n",
       "      <td>2018-02-22</td>\n",
       "      <td>1</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2018-02-23</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>115.5400</td>\n",
       "      <td>116.9900</td>\n",
       "      <td>0.012550</td>\n",
       "      <td>Information Technology</td>\n",
       "      <td>2018-02-27</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>115.5400</td>\n",
       "      <td>115.9300</td>\n",
       "      <td>0.003375</td>\n",
       "      <td>-0.009174</td>\n",
       "      <td>-0.009061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1513</td>\n",
       "      <td>21582</td>\n",
       "      <td>2019-11-14</td>\n",
       "      <td>1</td>\n",
       "      <td>Communication Services</td>\n",
       "      <td>2019-11-15</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>25.3500</td>\n",
       "      <td>26.5600</td>\n",
       "      <td>0.047732</td>\n",
       "      <td>Communication Services</td>\n",
       "      <td>2019-11-19</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>25.3500</td>\n",
       "      <td>26.7300</td>\n",
       "      <td>0.054438</td>\n",
       "      <td>0.006706</td>\n",
       "      <td>0.006401</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      company_id filing_date  n_trading             gics_sector        date  \\\n",
       "62         31816  2019-02-14          1             Health Care  2019-02-15   \n",
       "401         1070  2019-02-15          1               Materials  2019-02-19   \n",
       "1166       29618  2018-10-24          1  Information Technology  2018-10-25   \n",
       "1450        5060  2018-02-22          1  Information Technology  2018-02-23   \n",
       "1513       21582  2019-11-14          1  Communication Services  2019-11-15   \n",
       "\n",
       "      filing_released  n_calendar     close  close_next_n  percent_growth  \\\n",
       "62                  1           1   92.8511       94.9518        0.022624   \n",
       "401                 1           4   12.1166       12.9159        0.065967   \n",
       "1166                1           1  151.2189      154.9717        0.024817   \n",
       "1450                1           1  115.5400      116.9900        0.012550   \n",
       "1513                1           1   25.3500       26.5600        0.047732   \n",
       "\n",
       "     gics_sector_after_shift_days date_after_shift_days  \\\n",
       "62                    Health Care            2019-02-20   \n",
       "401                     Materials            2019-02-21   \n",
       "1166       Information Technology            2018-10-29   \n",
       "1450       Information Technology            2018-02-27   \n",
       "1513       Communication Services            2019-11-19   \n",
       "\n",
       "      filing_released_after_shift_days  n_calendar_after_shift_days  \\\n",
       "62                                   1                            6   \n",
       "401                                  1                            6   \n",
       "1166                                 1                            5   \n",
       "1450                                 1                            5   \n",
       "1513                                 1                            5   \n",
       "\n",
       "      close_after_shift_days  close_next_n_after_shift_days  \\\n",
       "62                   92.8511                        93.9761   \n",
       "401                  12.1166                        12.8468   \n",
       "1166                151.2189                       150.8436   \n",
       "1450                115.5400                       115.9300   \n",
       "1513                 25.3500                        26.7300   \n",
       "\n",
       "      percent_growth_after_shift_days  growth_increment  \\\n",
       "62                           0.012116         -0.010508   \n",
       "401                          0.060264         -0.005703   \n",
       "1166                        -0.002482         -0.027299   \n",
       "1450                         0.003375         -0.009174   \n",
       "1513                         0.054438          0.006706   \n",
       "\n",
       "      price_growth_during_shift_days  \n",
       "62                         -0.010276  \n",
       "401                        -0.005350  \n",
       "1166                       -0.026638  \n",
       "1450                       -0.009061  \n",
       "1513                        0.006401  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>company_id</th>\n",
       "      <th>filing_date</th>\n",
       "      <th>n_trading</th>\n",
       "      <th>gics_sector</th>\n",
       "      <th>date</th>\n",
       "      <th>filing_released</th>\n",
       "      <th>n_calendar</th>\n",
       "      <th>close</th>\n",
       "      <th>close_next_n</th>\n",
       "      <th>percent_growth</th>\n",
       "      <th>gics_sector_after_shift_days</th>\n",
       "      <th>date_after_shift_days</th>\n",
       "      <th>filing_released_after_shift_days</th>\n",
       "      <th>n_calendar_after_shift_days</th>\n",
       "      <th>close_after_shift_days</th>\n",
       "      <th>close_next_n_after_shift_days</th>\n",
       "      <th>percent_growth_after_shift_days</th>\n",
       "      <th>growth_increment</th>\n",
       "      <th>price_growth_during_shift_days</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>15071</td>\n",
       "      <td>2019-11-18</td>\n",
       "      <td>1</td>\n",
       "      <td>Industrials</td>\n",
       "      <td>2019-11-19</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>97.2207</td>\n",
       "      <td>99.0381</td>\n",
       "      <td>0.018694</td>\n",
       "      <td>Industrials</td>\n",
       "      <td>2019-11-21</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>97.2207</td>\n",
       "      <td>97.7299</td>\n",
       "      <td>0.005238</td>\n",
       "      <td>-0.013456</td>\n",
       "      <td>-0.013209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>2203</td>\n",
       "      <td>2019-02-08</td>\n",
       "      <td>1</td>\n",
       "      <td>Real Estate</td>\n",
       "      <td>2019-02-11</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>29.1661</td>\n",
       "      <td>28.9131</td>\n",
       "      <td>-0.008674</td>\n",
       "      <td>Real Estate</td>\n",
       "      <td>2019-02-13</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>29.1661</td>\n",
       "      <td>29.0931</td>\n",
       "      <td>-0.002503</td>\n",
       "      <td>0.006172</td>\n",
       "      <td>0.006226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>477</td>\n",
       "      <td>2019-02-26</td>\n",
       "      <td>1</td>\n",
       "      <td>Financials</td>\n",
       "      <td>2019-02-27</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>48.1646</td>\n",
       "      <td>48.2434</td>\n",
       "      <td>0.001636</td>\n",
       "      <td>Financials</td>\n",
       "      <td>2019-03-01</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>48.1646</td>\n",
       "      <td>48.4207</td>\n",
       "      <td>0.005317</td>\n",
       "      <td>0.003681</td>\n",
       "      <td>0.003675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>31206</td>\n",
       "      <td>2018-02-01</td>\n",
       "      <td>1</td>\n",
       "      <td>Communication Services</td>\n",
       "      <td>2018-02-02</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>30.8192</td>\n",
       "      <td>30.5788</td>\n",
       "      <td>-0.007800</td>\n",
       "      <td>Communication Services</td>\n",
       "      <td>2018-02-06</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>30.8192</td>\n",
       "      <td>36.3563</td>\n",
       "      <td>0.179664</td>\n",
       "      <td>0.187464</td>\n",
       "      <td>0.188938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>7345</td>\n",
       "      <td>2018-02-09</td>\n",
       "      <td>1</td>\n",
       "      <td>Financials</td>\n",
       "      <td>2018-02-12</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>89.2184</td>\n",
       "      <td>92.2919</td>\n",
       "      <td>0.034449</td>\n",
       "      <td>Financials</td>\n",
       "      <td>2018-02-14</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>89.2184</td>\n",
       "      <td>94.4152</td>\n",
       "      <td>0.058248</td>\n",
       "      <td>0.023799</td>\n",
       "      <td>0.023006</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    company_id filing_date  n_trading             gics_sector        date  \\\n",
       "0        15071  2019-11-18          1             Industrials  2019-11-19   \n",
       "8         2203  2019-02-08          1             Real Estate  2019-02-11   \n",
       "16         477  2019-02-26          1              Financials  2019-02-27   \n",
       "17       31206  2018-02-01          1  Communication Services  2018-02-02   \n",
       "19        7345  2018-02-09          1              Financials  2018-02-12   \n",
       "\n",
       "    filing_released  n_calendar    close  close_next_n  percent_growth  \\\n",
       "0                 0           1  97.2207       99.0381        0.018694   \n",
       "8                 0           3  29.1661       28.9131       -0.008674   \n",
       "16                0           1  48.1646       48.2434        0.001636   \n",
       "17                0           1  30.8192       30.5788       -0.007800   \n",
       "19                0           3  89.2184       92.2919        0.034449   \n",
       "\n",
       "   gics_sector_after_shift_days date_after_shift_days  \\\n",
       "0                   Industrials            2019-11-21   \n",
       "8                   Real Estate            2019-02-13   \n",
       "16                   Financials            2019-03-01   \n",
       "17       Communication Services            2018-02-06   \n",
       "19                   Financials            2018-02-14   \n",
       "\n",
       "    filing_released_after_shift_days  n_calendar_after_shift_days  \\\n",
       "0                                  0                            3   \n",
       "8                                  0                            5   \n",
       "16                                 0                            3   \n",
       "17                                 0                            5   \n",
       "19                                 0                            5   \n",
       "\n",
       "    close_after_shift_days  close_next_n_after_shift_days  \\\n",
       "0                  97.2207                        97.7299   \n",
       "8                  29.1661                        29.0931   \n",
       "16                 48.1646                        48.4207   \n",
       "17                 30.8192                        36.3563   \n",
       "19                 89.2184                        94.4152   \n",
       "\n",
       "    percent_growth_after_shift_days  growth_increment  \\\n",
       "0                          0.005238         -0.013456   \n",
       "8                         -0.002503          0.006172   \n",
       "16                         0.005317          0.003681   \n",
       "17                         0.179664          0.187464   \n",
       "19                         0.058248          0.023799   \n",
       "\n",
       "    price_growth_during_shift_days  \n",
       "0                        -0.013209  \n",
       "8                         0.006226  \n",
       "16                        0.003675  \n",
       "17                        0.188938  \n",
       "19                        0.023006  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shift = 2, Threshold = 0.01\n",
      "Filing Released: mean=-0.001, std=0.025, number_of_samples=257\n",
      "Ordinary Day: mean=0.002, std=0.026, number_of_samples=24315\n",
      "P-value: 0.072\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "th = 0.01\n",
    "\n",
    "# Extract samples for companies that published their reports\n",
    "filing_released_samples = list(df_merged[(df_merged['filing_released'] == 1) & (df_merged['percent_growth'] >= th)]['price_growth_during_shift_days'])\n",
    "display(df_merged[(df_merged['filing_released'] == 1) & (df_merged['percent_growth'] > th)].head())\n",
    "\n",
    "# Extract samples for companies that did not publish their reports\n",
    "ordinary_day_samples = list(df_merged[(df_merged['filing_released'] == 0)]['price_growth_during_shift_days'])\n",
    "display(df_merged[(df_merged['filing_released'] == 0)].head())\n",
    "\n",
    "# Get some basic statistics\n",
    "t_score, p_value = ttest_ind(filing_released_samples, ordinary_day_samples, equal_var=False)\n",
    "print(f'Shift = {shift}, Threshold = {th}')\n",
    "print(f'Filing Released: mean={np.mean(filing_released_samples):.3f}, std={np.std(filing_released_samples):.3f}, number_of_samples={len(filing_released_samples):d}')\n",
    "print(f'Ordinary Day: mean={np.mean(ordinary_day_samples):.3f}, std={np.std(ordinary_day_samples):.3f}, number_of_samples={len(ordinary_day_samples):d}')\n",
    "print(f'P-value: {p_value:.3f}')\n",
    "print('\\n')\n",
    "\n",
    "# Make a histogram\n",
    "print_hist(filing_released_samples, ordinary_day_samples, 'Price Growth after Financial Statement was Released')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Streamlining\n",
    "Let's perform this analysis for a range of different shifts and price growth thresholds to find out whether there is a combination indicating significant diffirence between price behaviour among companies that released their financial statement and those that didn't."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shift = 1, Threshold = 0.005\n",
      "Filing Released: mean=0.001, std=0.017, number_of_samples=372\n",
      "Ordinary Day: mean=0.001, std=0.019, number_of_samples=24334\n",
      "P-value: 0.851\n",
      "\n",
      "\n",
      "Shift = 1, Threshold = 0.01\n",
      "Filing Released: mean=0.001, std=0.018, number_of_samples=257\n",
      "Ordinary Day: mean=0.001, std=0.019, number_of_samples=24334\n",
      "P-value: 0.918\n",
      "\n",
      "\n",
      "Shift = 1, Threshold = 0.015\n",
      "Filing Released: mean=0.001, std=0.019, number_of_samples=156\n",
      "Ordinary Day: mean=0.001, std=0.019, number_of_samples=24334\n",
      "P-value: 0.823\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.005\n",
      "Filing Released: mean=-0.000, std=0.023, number_of_samples=372\n",
      "Ordinary Day: mean=0.002, std=0.026, number_of_samples=24315\n",
      "P-value: 0.031\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.01\n",
      "Filing Released: mean=-0.001, std=0.025, number_of_samples=257\n",
      "Ordinary Day: mean=0.002, std=0.026, number_of_samples=24315\n",
      "P-value: 0.072\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.015\n",
      "Filing Released: mean=-0.002, std=0.026, number_of_samples=156\n",
      "Ordinary Day: mean=0.002, std=0.026, number_of_samples=24315\n",
      "P-value: 0.054\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.005\n",
      "Filing Released: mean=-0.001, std=0.029, number_of_samples=371\n",
      "Ordinary Day: mean=0.003, std=0.032, number_of_samples=24212\n",
      "P-value: 0.013\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.01\n",
      "Filing Released: mean=-0.002, std=0.030, number_of_samples=257\n",
      "Ordinary Day: mean=0.003, std=0.032, number_of_samples=24212\n",
      "P-value: 0.008\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.015\n",
      "Filing Released: mean=-0.004, std=0.032, number_of_samples=156\n",
      "Ordinary Day: mean=0.003, std=0.032, number_of_samples=24212\n",
      "P-value: 0.020\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.005\n",
      "Filing Released: mean=-0.005, std=0.034, number_of_samples=296\n",
      "Ordinary Day: mean=0.002, std=0.038, number_of_samples=18859\n",
      "P-value: 0.002\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.01\n",
      "Filing Released: mean=-0.006, std=0.034, number_of_samples=212\n",
      "Ordinary Day: mean=0.002, std=0.038, number_of_samples=18859\n",
      "P-value: 0.001\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.015\n",
      "Filing Released: mean=-0.007, std=0.037, number_of_samples=130\n",
      "Ordinary Day: mean=0.002, std=0.038, number_of_samples=18859\n",
      "P-value: 0.010\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Wrap the process discribed above in a function for conveninet implementation\n",
    "def evaluate(df, shift, th, display=True, title=''):\n",
    "    df_ = df.copy()\n",
    "    df_shift = df_.copy()\n",
    "    df_shift['n_trading'] = df_shift['n_trading'] - shift\n",
    "    index_columns = ['company_id', 'filing_date', 'n_trading']\n",
    "    df_merged = df_.set_index(index_columns).join(\n",
    "        df_shift.set_index(index_columns), \n",
    "        how='inner',\n",
    "        rsuffix='_after_shift_days'\n",
    "    ).reset_index()\n",
    "    df_merged = df_merged[df_merged['n_trading'] == 1]\n",
    "    df_merged['growth_increment'] = df_merged['percent_growth_after_shift_days'] - df_merged['percent_growth']\n",
    "    df_merged['price_growth_during_n_days'] = df_merged['close_next_n_after_shift_days'] / df_merged['close_next_n'] - 1\n",
    "\n",
    "    filing_released_samples = list(df_merged[(df_merged['filing_released'] == 1) & (df_merged['percent_growth'] > th)]['price_growth_during_n_days'])\n",
    "    ordinary_day_samples = list(df_merged[(df_merged['filing_released'] == 0)]['price_growth_during_n_days'])\n",
    "    \n",
    "    if display:\n",
    "        print_hist(filing_released_samples, ordinary_day_samples, title=title)\n",
    "        \n",
    "    t_score, p_value = ttest_ind(filing_released_samples, ordinary_day_samples, equal_var=False)\n",
    "    \n",
    "    print(f'Shift = {shift}, Threshold = {th}')\n",
    "    print(f'Filing Released: mean={np.mean(filing_released_samples):.3f}, std={np.std(filing_released_samples):.3f}, number_of_samples={len(filing_released_samples):d}')\n",
    "    print(f'Ordinary Day: mean={np.mean(ordinary_day_samples):.3f}, std={np.std(ordinary_day_samples):.3f}, number_of_samples={len(ordinary_day_samples):d}')\n",
    "    print(f'P-value: {p_value:.3f}')\n",
    "    print('\\n')\n",
    "    \n",
    "# Define parameter space\n",
    "shifts = [1, 2, 3, 4]\n",
    "thresholds = [0.005, 0.01, 0.015]\n",
    "\n",
    "# Iterate through the parameters and print out statistics\n",
    "for shift in shifts:\n",
    "    for th in thresholds:\n",
    "        evaluate(df=df, shift=shift, th=th, display=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Observation\n",
    "The most promising run (lowest p-value) was at shift=4 and threshold=0.01 (or 1% price growth during the first day after the filing). Let's take a closer look at it's histogram."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shift = 4, Threshold = 0.01\n",
      "Filing Released: mean=-0.006, std=0.034, number_of_samples=212\n",
      "Ordinary Day: mean=0.002, std=0.038, number_of_samples=18859\n",
      "P-value: 0.001\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "th = 0.01\n",
    "shift = 4\n",
    "evaluate(df=df, shift=shift, th=th, display=True, title=f'Price Growth after Financial Statement was Released (shift={shift}, th={th})')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Indeed,</b> we can see that the historgram for the 'Filing Released' companies is shifted towards left comparad to the histogram for companies on an ordinary day. Keeping in mind that 'Filing Released' data has been conditioned on at least 1% price growth immediately after financial statement release. This leads us to conslusion that in the timerage of this study, between 2018 and 2020, companies on average had a tendency to correct their immediate growth within 4 days after their financial statement release comparad to other companies under (assumignly) normal market conditions.\n",
    "\n",
    "<b>Next</b>, let's perform the same analysis segmenting our sampling space to the industry sectors. In other words, we will compare companies that released their financial statements only with companies that come from the same sector. This will allow us to reduce the number of random factors affecting our results. For example, if a given sector is experiencing strong up-trend, we'd want to compare companies that released financial statements as part of this up-trend with companies that didn't release their statements as part of this same up-trend, excluding companies form declining or drifting industries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Computing statistics for Informaion Technology sector\n",
      "\n",
      "Shift = 1, Threshold = 0.005\n",
      "Filing Released: mean=0.001, std=0.014, number_of_samples=47\n",
      "Ordinary Day: mean=0.001, std=0.020, number_of_samples=5811\n",
      "P-value: 0.820\n",
      "\n",
      "\n",
      "Shift = 1, Threshold = 0.01\n",
      "Filing Released: mean=0.002, std=0.015, number_of_samples=29\n",
      "Ordinary Day: mean=0.001, std=0.020, number_of_samples=5811\n",
      "P-value: 0.621\n",
      "\n",
      "\n",
      "Shift = 1, Threshold = 0.015\n",
      "Filing Released: mean=-0.001, std=0.015, number_of_samples=19\n",
      "Ordinary Day: mean=0.001, std=0.020, number_of_samples=5811\n",
      "P-value: 0.695\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.005\n",
      "Filing Released: mean=0.001, std=0.021, number_of_samples=47\n",
      "Ordinary Day: mean=0.002, std=0.029, number_of_samples=5802\n",
      "P-value: 0.770\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.01\n",
      "Filing Released: mean=0.003, std=0.022, number_of_samples=29\n",
      "Ordinary Day: mean=0.002, std=0.029, number_of_samples=5802\n",
      "P-value: 0.802\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.015\n",
      "Filing Released: mean=-0.001, std=0.018, number_of_samples=19\n",
      "Ordinary Day: mean=0.002, std=0.029, number_of_samples=5802\n",
      "P-value: 0.512\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.005\n",
      "Filing Released: mean=0.002, std=0.025, number_of_samples=47\n",
      "Ordinary Day: mean=0.002, std=0.035, number_of_samples=5755\n",
      "P-value: 0.977\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.01\n",
      "Filing Released: mean=0.004, std=0.026, number_of_samples=29\n",
      "Ordinary Day: mean=0.002, std=0.035, number_of_samples=5755\n",
      "P-value: 0.639\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.015\n",
      "Filing Released: mean=0.003, std=0.025, number_of_samples=19\n",
      "Ordinary Day: mean=0.002, std=0.035, number_of_samples=5755\n",
      "P-value: 0.930\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.005\n",
      "Filing Released: mean=-0.002, std=0.031, number_of_samples=41\n",
      "Ordinary Day: mean=0.002, std=0.040, number_of_samples=4736\n",
      "P-value: 0.453\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.01\n",
      "Filing Released: mean=-0.004, std=0.031, number_of_samples=25\n",
      "Ordinary Day: mean=0.002, std=0.040, number_of_samples=4736\n",
      "P-value: 0.340\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.015\n",
      "Filing Released: mean=-0.004, std=0.035, number_of_samples=16\n",
      "Ordinary Day: mean=0.002, std=0.040, number_of_samples=4736\n",
      "P-value: 0.540\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_it = df[df['gics_sector'] == 'Information Technology']\n",
    "print('Computing statistics for Informaion Technology sector\\n')\n",
    "shifts = [1, 2, 3, 4]\n",
    "thresholds = [0.005, 0.01, 0.015]\n",
    "for shift in shifts:\n",
    "    for th in thresholds:\n",
    "        evaluate(df=df_it, shift=shift, th=th, display=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Computing statistics for Health Care sector\n",
      "\n",
      "Shift = 1, Threshold = 0.005\n",
      "Filing Released: mean=0.004, std=0.026, number_of_samples=48\n",
      "Ordinary Day: mean=0.002, std=0.019, number_of_samples=3004\n",
      "P-value: 0.457\n",
      "\n",
      "\n",
      "Shift = 1, Threshold = 0.01\n",
      "Filing Released: mean=0.005, std=0.027, number_of_samples=30\n",
      "Ordinary Day: mean=0.002, std=0.019, number_of_samples=3004\n",
      "P-value: 0.480\n",
      "\n",
      "\n",
      "Shift = 1, Threshold = 0.015\n",
      "Filing Released: mean=0.013, std=0.022, number_of_samples=18\n",
      "Ordinary Day: mean=0.002, std=0.019, number_of_samples=3004\n",
      "P-value: 0.056\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.005\n",
      "Filing Released: mean=0.000, std=0.031, number_of_samples=48\n",
      "Ordinary Day: mean=0.004, std=0.025, number_of_samples=2994\n",
      "P-value: 0.376\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.01\n",
      "Filing Released: mean=-0.001, std=0.032, number_of_samples=30\n",
      "Ordinary Day: mean=0.004, std=0.025, number_of_samples=2994\n",
      "P-value: 0.332\n",
      "\n",
      "\n",
      "Shift = 2, Threshold = 0.015\n",
      "Filing Released: mean=-0.001, std=0.036, number_of_samples=18\n",
      "Ordinary Day: mean=0.004, std=0.025, number_of_samples=2994\n",
      "P-value: 0.546\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.005\n",
      "Filing Released: mean=-0.003, std=0.037, number_of_samples=47\n",
      "Ordinary Day: mean=0.005, std=0.032, number_of_samples=2941\n",
      "P-value: 0.150\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.01\n",
      "Filing Released: mean=-0.006, std=0.034, number_of_samples=30\n",
      "Ordinary Day: mean=0.005, std=0.032, number_of_samples=2941\n",
      "P-value: 0.100\n",
      "\n",
      "\n",
      "Shift = 3, Threshold = 0.015\n",
      "Filing Released: mean=0.000, std=0.039, number_of_samples=18\n",
      "Ordinary Day: mean=0.005, std=0.032, number_of_samples=2941\n",
      "P-value: 0.629\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.005\n",
      "Filing Released: mean=-0.006, std=0.048, number_of_samples=31\n",
      "Ordinary Day: mean=0.001, std=0.040, number_of_samples=1926\n",
      "P-value: 0.408\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.01\n",
      "Filing Released: mean=-0.007, std=0.043, number_of_samples=22\n",
      "Ordinary Day: mean=0.001, std=0.040, number_of_samples=1926\n",
      "P-value: 0.405\n",
      "\n",
      "\n",
      "Shift = 4, Threshold = 0.015\n",
      "Filing Released: mean=0.001, std=0.052, number_of_samples=13\n",
      "Ordinary Day: mean=0.001, std=0.040, number_of_samples=1926\n",
      "P-value: 0.985\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df_it = df[df['gics_sector'] == 'Health Care']\n",
    "print('Computing statistics for Health Care sector\\n')\n",
    "for shift in shifts:\n",
    "    for th in thresholds:\n",
    "        evaluate(df=df_it, shift=shift, th=th, display=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "\n",
    "The goal of this study was to briefly assess whether the effect of financial statement release has some correlation with the stock price behaviour and to determine whether it's worth the effort to explore this topic in depth.\n",
    "\n",
    "While some statistically significant observations were noted, the results overall show rather week correlation between the event of financial statement release and the stock behaviour."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:root] *",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}