21_hypothesis_testing_normal.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# ICS 434: DATA SCIENCE FUNDAMENTALS\n",
    "\n",
    "## Hypothesis Testing\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import scipy as sp\n",
    "\n",
    "np.random.seed(129)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "### Hypothesis Testing\n",
    "\n",
    "* Definitions:\n",
    "   * A statistic is the piece of information computed directly on a sample\n",
    "   * A parameter is the opposite of the statistic and applies to the complete population\n",
    "\n",
    "* Hypothesis is an assumption about a population parameter\n",
    "\n",
    "* The goal of a hypothesis test is to select between two competing hypotheses\n",
    "\n",
    "* The difference between parametric a non-parametric methods:\n",
    "  * Parametric methods make assumptions about the parameters and the underlying distributions\n",
    "  * Non-parametric methods do not make any assumptions about underlying params and their distributions\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "### Hypothesis Testing Questions\n",
    "\n",
    "* In hypothesis testing, we are typically interested in:\n",
    "    1. A measure of the difference between two hypotheses.\n",
    "        * This is referred to as the `effect size`.\n",
    "        * Ex. Economists have shown that universal basic income makes people happier.\n",
    "            * The results were statistically significant.\n",
    "            * But how much happier are they? If participants are 1% happier, is it worth spending \\$43 billion (expected cost in Canada) on the universal basic income?\n",
    "\n",
    "    2. Probabilities associated with the measured difference \n",
    "      * How variable is the measured or inferred difference?\n",
    "\n",
    "        * How likely is it that the difference is due to chance alone?\n",
    "\n",
    "        * This is referred to as `p-value.`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Data in Hypothesis Testing\n",
    "\n",
    "\n",
    "* Can involve data from either one or two samples. For instance:\n",
    "    * Comparing a single sample to a population with a known mean.  \n",
    "\n",
    "        * E.g., you use a new math curriculum on a *sample* of randomly selected students and compare the results to the national average. \n",
    "        * Does your curriculum improve math exam scores (compared to the national average)?\n",
    " \n",
    "    * Comparing two samples to determine whether there is a difference between them.\n",
    "        * E.g., you teach your math curriculum to a group of students in the morning and another group of students in the afternoon.\n",
    "        * Is there a difference between learning math in the morning or the afternoon?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Terminology\n",
    "\n",
    "* We are typically testing two hypotheses\n",
    "\n",
    "  * The first hypothesis assumes that there is no change in our system \n",
    "    * Any changes that we observe are merely due to natural variance in the population\n",
    "    * This is referred to as the null hypothesis ($H_0$)\n",
    "  * The second hypothesis assumes the differences we are observing are not likely to arise due to chance alone\n",
    "    * By \"not likely,\" we mean something that would probabilistically rare in occurrence\n",
    "    * This is referred to as the alternative hypothesis ($H_a$)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Terminology -- Cont'd\n",
    "\n",
    "* The alternative hypothesis can be:\n",
    "\n",
    "1. Single-tailed\n",
    "  * Our new method improves the average score compared to the national average<br>\n",
    "    $\\mu_{new\\_method} > \\mu_{national\\_average}$\n",
    "   \n",
    "  * A new medication decreases cholesterol levels  \n",
    "    $\\mu_{cholesterol\\_treatment} < \\mu_{cholesterol\\_control}$\n",
    "\n",
    "2. Two-tailed\n",
    "  * The vlaues must be different from a given value\n",
    "   * Ex. Rate of prescription of opioids in Hawaii is different from the national average. This implied<br>\n",
    "   $\\mu_{opioids\\_hawaii} < \\mu_{opioids\\_nationally}$ or\n",
    "   $\\mu_{opioids\\_hawaii} > \\mu_{opioids\\_nationally}$\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "### About the Hypothesis Tests\n",
    "\n",
    "* Hypotheses ($H_0$ and $H_a$) are mutually exclusive.\n",
    "  * Ex. There is a difference or there is no difference.\n",
    "  * But not: the difference is greater than 0 or greater than 10. \n",
    "\n",
    "* A null hypothesis can never be proven to be true.\n",
    "  * Failing to prove that the alternative is true only makes the null plausible. \n",
    "  * We say that we fail to reject the null hypothesis or that the study did not find sufficient evidence to reject the null hypothesis.\n",
    "\n",
    "* We will say that the results are statistically significant when they are unlikely to be due to chance alone."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Example: Data from a Theoretical Distribution\n",
    "\n",
    "* Suppose the data comes from two normal distributions\n",
    "  * The distributions have means $\\mu_1= 232$ and $\\mu_2 = 219$\n",
    "  * The standard deviations are the same for both distributions ($\\sigma = 12$)\n",
    "  \n",
    "* These could perhaps your app average ad time in minutes viewed daily on YouTube versus Vimeo\n",
    "   * We have 90 days worth of viewing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "plt.figure(figsize=(12, 6))\n",
    "youtube = np.random.normal(232, 12, 90)\n",
    "vimeo = np.random.normal(219, 12, 90)\n",
    "\n",
    "kde_1 = sp.stats.gaussian_kde(youtube, bw_method=0.51)\n",
    "kde_2 = sp.stats.gaussian_kde(vimeo, bw_method=0.52)\n",
    "\n",
    "x_values = np.arange(160, 280, 0.25)\n",
    "# Estimate kernel density for the support (x-axis)\n",
    "x_densities_1 = kde_1.evaluate(x_values)\n",
    "x_densities_2 = kde_2.evaluate(x_values)\n",
    "plt.plot(x_values, x_densities_1)\n",
    "plt.plot(x_values, x_densities_2);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Assessing the Variability of Our Statistic: Different Samples\n",
    "\n",
    "\n",
    "* The difference between the means is a useful statistic to assess the difference between both samples\n",
    "  * We call a statistic computed on the observed data the _observed statistic_\n",
    "\n",
    "* Given that the data consists of random samples, collecting new data will undoubtedly lead to a change in observed statistic\n",
    "\n",
    "* We need a distribution of the difference between the two samples to compare our statistic\n",
    "  * This is the distribution of the test statistic under the null hypothesis\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11.159994107248792"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(youtube) - np.mean(vimeo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Assessing the Variability of Our Statistic: Different Samples -- Cont'd\n",
    "\n",
    "* We will start with an example where the samples are clearly (visually) different\n",
    "  * To do so, we generate a dataset with $\\sigma=2$\n",
    "\n",
    "<center><img src=\"https://www.dropbox.com/s/hozq22hz2fyjrg8/two_dists.png?dl=1\"></center>\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 3.,  8., 14., 10., 14., 19., 10.,  5.,  2.,  3.,  2.]),\n",
       " array([215.16245757, 216.00590128, 216.84934498, 217.69278869,\n",
       "        218.53623239, 219.3796761 , 220.2231198 , 221.06656351,\n",
       "        221.91000722, 222.75345092, 223.59689463, 224.44033833]),\n",
       " <BarContainer object of 11 artists>)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgAAAAESCAYAAACLjm33AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8WgzjOAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAYBElEQVR4nO3df3BU1d3H8c8CYbNoEljIDzIEpJaiLYiVumlG28HKBDIOlRb+MP1D7DDa2gWGpi3TdAoU7BhbHXGcBpg6FXQ6xNrMQK21WIqyTAXMoKaVTqEEMWI3CYUrWbIkmwDn+cOHfZ6YCNndu8km5/2auTPs3XPP/d453M0ne2/O9RhjjAAAgFVGDXUBAABg8BEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAAC40Z6gI+6fLlywqHw8rJyZHH4xnqcgAAGDaMMTp//ryKi4s1atTVf8fPuAAQDodVUlIy1GUAADBsnTp1SlOmTLlqm4wLADk5OZI+Lj43N3eIqwEAYPiIRCIqKSmJ/yy9mowLAFe+9s/NzSUAAACQhIFcQucmQAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACyUUACoqanR7bffrpycHBUUFGjx4sU6duxYrzZdXV0KBoOaOHGirr/+ei1ZskRtbW2uFg0AAFKTUAAIhUIKBoM6dOiQ9uzZo56eHpWXlysajcbbfP/739cf//hH/f73v1coFFI4HNY3v/lN1wsHAADJ8xhjTLIb//e//1VBQYFCoZC++tWvqr29Xfn5+dqxY4eWLl0qSTp69KhuvvlmHTx4UF/+8pf79BGLxRSLxeKvr0xi0N7ezjwAAOCiuro6OY6Tcj9+v1+VlZUuVAS3RSIR5eXlDehnaEoTAbW3t0v6+D+DJL311lvq6enR/Pnz421uuukmTZ069VMDQE1NjTZs2JBKGQCAAXAcR059vfw+X/J9dHZK//sLHoa3pAPA5cuXtXr1at1xxx2aNWuWJKm1tVVjx47V+PHje7UtLCxUa2trv/1UV1erqqoq/vrKNwAAAPf5fT4FA4Gkt69taHCxGgylpANAMBjUkSNH9Le//S2lArxer7xeb0p9AACAxCT1Z4ArVqzQyy+/rNdff73X04aKiorU3d2tc+fO9Wrf1tamoqKilAoFAADuSSgAGGO0YsUK7dy5U6+99pqmT5/e6/25c+cqKytLe/fuja87duyYPvjgA5WVlblTMQAASFlClwCCwaB27NihP/zhD8rJyYlf18/Ly5PP51NeXp6WL1+uqqoq+f1+5ebmauXKlSorK+v3BkAAADA0EgoAW7ZskSTNmzev1/pt27bpgQcekCRt2rRJo0aN0pIlSxSLxbRgwQJt3rzZlWIBAIA7EgoAA5kyIDs7W7W1taqtrU26KAAAkF48CwAAAAsRAAAAsFBKMwECbk0tKjG9KAAMJgIAUuI4jurrHfl8/pT66ex0mF0UAAYRAQAp8/n8CgSCKfXR0MBNowAwmLgHAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALJRwA9u/fr0WLFqm4uFgej0e7du3q9f4DDzwgj8fTa1m4cKFb9QIAABckHACi0ajmzJmj2traT22zcOFCtbS0xJe6urqUigQAAO4ak+gGFRUVqqiouGobr9eroqKiAfUXi8UUi8XiryORSKIlAQCABKXlHoB9+/apoKBAM2fO1MMPP6yzZ89+atuamhrl5eXFl5KSknSUBAAA/h/XA8DChQv1/PPPa+/evfrFL36hUCikiooKXbp0qd/21dXVam9vjy+nTp1yuyQAAPAJCV8CuJb77rsv/u/Zs2frlltu0Y033qh9+/bp7rvv7tPe6/XK6/W6XQYAALiKtP8Z4Gc+8xlNmjRJTU1N6d4VAAAYoLQHgA8//FBnz57V5MmT070rAAAwQAlfAujo6Oj12/zJkyfV2Ngov98vv9+vDRs2aMmSJSoqKtKJEye0Zs0affazn9WCBQtcLRwAACQv4QBw+PBh3XXXXfHXVVVVkqRly5Zpy5Yt+sc//qHnnntO586dU3FxscrLy/XII49wnR8AgAyScACYN2+ejDGf+v6rr76aUkEAACD9eBYAAAAWIgAAAGAh1+cBwNXV1dXJcRxX+vL7/aqsrBzSWkKhkMLhGSn1AcAebn0Gpvr5BwLAoHMcR/X1jnw+f0r9dHY6Wrp06GtparqgrKzu1AoBYA3HceTU18vv8yXfR2enUv4ABAFgKPh8fgUCwZT6aGj49KcxDmYtzc0hV+oAYA+/z6dgIJD09rUNDS5WYy/uAQAAwEIEAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEJjhroAAMDwcTgcVjQUSnr7UCikGeGwixUhWQQAAMCARXt6NKGxUeroSGr7C01N6s7KcrcoJIUAAABISG5WloKBQFLbhpqbXa4GyeIeAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQkwEBAAYVlKdjvgKv9+vyspKFyoanggAAIBhJdXpiCXJ6eyUli51r6hhiAAAABh2UpmOWJJqGxpcrGZ44h4AAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQgkHgP3792vRokUqLi6Wx+PRrl27er1vjNG6des0efJk+Xw+zZ8/X8ePH3erXgAA4IKEA0A0GtWcOXNUW1vb7/u//OUv9fTTT2vr1q168803dd1112nBggXq6upKuVgAAOCOhCcCqqioUEVFRb/vGWP01FNP6ac//anuvfdeSdLzzz+vwsJC7dq1S/fdd1+fbWKxmGKxWPx1JBJJtCQrhcOHFQpFU+ojFAopHJ7hUkWpceN4JKb2BICBcnUmwJMnT6q1tVXz58+Pr8vLy1NpaakOHjzYbwCoqanRhg0b3CzDCj09UTU2TkhlJkw1NV1QVla3e0WlwI3j6ex0bJ/ZEwAGzNUA0NraKkkqLCzstb6wsDD+3idVV1erqqoq/joSiaikpMTNskasrKxcBQLBpLdvbk79YRpuSvV4Ghr6vywFAOhryJ8F4PV65fV6h7oMAACs4uqfARYVFUmS2traeq1va2uLvwcAAIaeqwFg+vTpKioq0t69e+PrIpGI3nzzTZWVlbm5KwAAkIKELwF0dHSoqakp/vrkyZNqbGyU3+/X1KlTtXr1av385z/XjBkzNH36dK1du1bFxcVavHixm3UDAIAUJBwADh8+rLvuuiv++soNfMuWLdP27du1Zs0aRaNRPfTQQzp37pzuvPNO7d69W9nZ2e5VDQAAUpJwAJg3b56MMZ/6vsfj0caNG7Vx48aUCgMAAOnDswAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALBQwk8DtFldXZ0cx0mpj1AopHB4hksVAbCFW58/M8JhlyrCcEcASIDjOKqvd+Tz+ZPuo6npgrKyul2sCoANHMeRU18vv8+XdB8XmprUnZXlYlUYzggACfL5/AoEgklv39wccrEaADbx+3wKBgJJbx9qbnaxGgx33AMAAICFCAAAAFiIAAAAgIUIAAAAWIgAAACAhQgAAABYiAAAAICFCAAAAFiIiYAAIM2YxnfkcWNMJcnv96uystKFihJHAACANGMa35HHjTF1OjulpUtdrCoxBAAAGARM4zvypDqmtQ0NLlaTOO4BAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEKuB4Cf/exn8ng8vZabbrrJ7d0AAIAUjElHp1/4whf017/+9f92MiYtuwEAAElKy0/mMWPGqKioaEBtY7GYYrFY/HUkEklHSbBAOHxYoVA05X78fr8qKytdqAgjRV1dnRzHSXr7UCikGeGwixUBqUtLADh+/LiKi4uVnZ2tsrIy1dTUaOrUqf22ramp0YYNG9JRBizT0xNVY+MEdXQk30dnp6OlS92rCSOD4zhy6uvl9/mS2v5CU5O6s7JcrgpIjesBoLS0VNu3b9fMmTPV0tKiDRs26Ctf+YqOHDminJycPu2rq6tVVVUVfx2JRFRSUuJ2WbBEVlauAoFg0ts3NNS6WA1GEr/Pp2AgkNS2oeZml6sBUud6AKioqIj/+5ZbblFpaammTZumF198UcuXL+/T3uv1yuv1ul0GAAC4irT/GeD48eP1uc99Tk1NTeneFQAAGKC0B4COjg6dOHFCkydPTveuAADAALkeAH74wx8qFArp/fff14EDB/SNb3xDo0eP5q5qAAAyiOv3AHz44YeqrKzU2bNnlZ+frzvvvFOHDh1Sfn6+27sCAABJcj0AvPDCC253CQAAXMazAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsBABAAAACxEAAACwEAEAAAALEQAAALAQAQAAAAsRAAAAsJDrTwPMRHV1dXIcJ+V+QqGQwuEZLlSEkc6t/3N+v1+VlZUuVGQvN8YiFAppRjjsUkVAZrAiADiOo/p6Rz6fP6V+mpouKCur26WqMJK58X+us9PR0qUuFmUpx3Hk1NfL7/Ml3ceFpiZ1Z2W5WBUw9KwIAJLk8/kVCART6qO5OeRSNbBBqv/nGhpqXazGbn6fT8FAIOntQ83NLlYDZAbuAQAAwEIEAAAALEQAAADAQgQAAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALGTNREDAQITDhxUKRVPux41po92qhemEgb4Oh8OKhpKf3G0kTA9NAAD+n56eqBobJ6ijI7V+3Jg22o1amE4Y6F+0p0cTGhuV7Ak2EqaHJgAAn5CVlZsx00anWgvTCQOfLjcrK+kpokfC9NDcAwAAgIUIAAAAWIgAAACAhQgAAABYiAAAAICFCAAAAFiIAAAAgIUIAAAAWIiJgABkrLq6OjmOk1IfI2HKViAdCAAAMpbjOHLq6+X3+ZLuYyRM2QqkAwEAQEbz+3xJT9cqjYwpW4F04B4AAAAsRAAAAMBCBAAAACxEAAAAwEIEAAAALEQAAADAQmkLALW1tbrhhhuUnZ2t0tJSNTQ0pGtXAAAgQWkJAL/73e9UVVWl9evX6+2339acOXO0YMECnT59Oh27AwAACUrLREBPPvmkHnzwQX3729+WJG3dulV/+tOf9Oyzz+rHP/5xr7axWEyxWCz+ur29XZIUiURcq6ezs1Pnz3+kN954IqV+urraJZ1JqR83+sikWjiezK6lq+sjdXZOcPV8GkydnZ366Px5PfHGG0n30d7VpTPSkPaRCTVkSh+ZUIMbfbhRw0ddXZrQ2enq+XmlL2PMtRsbl8ViMTN69Gizc+fOXuvvv/9+8/Wvf71P+/Xr1xtJLCwsLCwsLC4tp06duubPa9e/AThz5owuXbqkwsLCXusLCwt19OjRPu2rq6tVVVUVf3358mU5jqOJEyfK4/EkXUckElFJSYlOnTql3NzcpPuB+xibzMXYZC7GJnNl0tgYY3T+/HkVFxdfs+2QPwvA6/XK6/X2Wjd+/HjX+s/NzR3yAUH/GJvMxdhkLsYmc2XK2OTl5Q2ones3AU6aNEmjR49WW1tbr/VtbW0qKipye3cAACAJrgeAsWPHau7cudq7d2983eXLl7V3716VlZW5vTsAAJCEtFwCqKqq0rJly/SlL31JgUBATz31lKLRaPyvAgaD1+vV+vXr+1xewNBjbDIXY5O5GJvMNVzHxmPMQP5WIHG/+tWv9Pjjj6u1tVW33nqrnn76aZWWlqZjVwAAIEFpCwAAACBz8SwAAAAsRAAAAMBCBAAAACxEAAAAwELDKgDU1NTo9ttvV05OjgoKCrR48WIdO3asV5tf//rXmjdvnnJzc+XxeHTu3Lk+/dxwww3yeDy9lscee2yQjmJkutbYOI6jlStXaubMmfL5fJo6dapWrVoVf/jTFR988IHuuecejRs3TgUFBfrRj36kixcvDvbhjChujc0nzxmPx6MXXnhhsA9nRBnIZ9p3vvMd3XjjjfL5fMrPz9e9997bZ1p1zhv3uTU2mXzeDKsAEAqFFAwGdejQIe3Zs0c9PT0qLy9XNBqNt7lw4YIWLlyon/zkJ1fta+PGjWppaYkvK1euTHf5I9q1xiYcDiscDuuJJ57QkSNHtH37du3evVvLly+P93Hp0iXdc8896u7u1oEDB/Tcc89p+/btWrdu3VAd1ojgxthcsW3btl7nzeLFiwf5aEaWgXymzZ07V9u2bdO//vUvvfrqqzLGqLy8XJcuXZLEeZMubozNFRl73qT8+L8hdPr0aSPJhEKhPu+9/vrrRpL56KOP+rw3bdo0s2nTpvQXaLGrjc0VL774ohk7dqzp6ekxxhjzyiuvmFGjRpnW1tZ4my1btpjc3FwTi8XSXrMtkhkbY4yR1Ocpn3DXQMbm73//u5FkmpqajDGcN4MlmbExJrPPm2H1DcAnXfmK0u/3J7ztY489pokTJ+qLX/yiHn/8cb4uc9lAxqa9vV25ubkaM+bjCSkPHjyo2bNn93qS5IIFCxSJRPTPf/4zvQVbJJmxuSIYDGrSpEkKBAJ69tlnB/bMcQzYtcYmGo1q27Ztmj59ukpKSiRx3gyWZMbmikw9b4b8aYDJunz5slavXq077rhDs2bNSmjbVatW6bbbbpPf79eBAwdUXV2tlpYWPfnkk2mq1i4DGZszZ87okUce0UMPPRRf19ra2u9jpK+8h9QlOzbSx5fNvva1r2ncuHH6y1/+ou9973vq6OjQqlWrBqP0Ee9qY7N582atWbNG0WhUM2fO1J49ezR27FhJnDeDIdmxkTL8vBnqryCS9d3vftdMmzbNnDp1qt/3r3YJ4JN+85vfmDFjxpiuri6Xq7TTtcamvb3dBAIBs3DhQtPd3R1f/+CDD5ry8vJebaPRqJFkXnnllbTWbItkx6Y/a9euNVOmTElHmVa62ticO3fO/Pvf/zahUMgsWrTI3Hbbbaazs9MYw3kzGJIdm/5k0nkzLANAMBg0U6ZMMe+9996ntkkkABw5csRIMkePHnWxSjtda2wikYgpKyszd999d5+TZO3atWbOnDm91r333ntGknn77bfTVbI1Uhmb/rz88stGEsHZBQP5TLsiFouZcePGmR07dhhjOG/SLZWx6U8mnTfD6h4AY4xWrFihnTt36rXXXtP06dNd6bexsVGjRo1SQUGBK/3ZaCBjE4lEVF5errFjx+qll15SdnZ2r/fLysr07rvv6vTp0/F1e/bsUW5urj7/+c+n/RhGKjfGpj+NjY2aMGHCsHsCWiZJ5jPNfPyLm2KxmCTOm3RxY2z6k1HnzZBFjyQ8/PDDJi8vz+zbt8+0tLTElwsXLsTbtLS0mHfeecc888wzRpLZv3+/eeedd8zZs2eNMcYcOHDAbNq0yTQ2NpoTJ06Y3/72tyY/P9/cf//9Q3VYI8K1xqa9vd2Ulpaa2bNnm6ampl5tLl68aIwx5uLFi2bWrFmmvLzcNDY2mt27d5v8/HxTXV09lIc27LkxNi+99JJ55plnzLvvvmuOHz9uNm/ebMaNG2fWrVs3lIc27F1rbE6cOGEeffRRc/jwYdPc3GzeeOMNs2jRIuP3+01bW5sxhvMmXdwYm0w/b4ZVAJDU77Jt27Z4m/Xr11+1zVtvvWVKS0tNXl6eyc7ONjfffLN59NFHM+LrmOHsWmNz5ZJMf8vJkyfj/bz//vumoqLC+Hw+M2nSJPODH/yg15+iIXFujM2f//xnc+utt5rrr7/eXHfddWbOnDlm69at5tKlS0N3YCPAtcbmP//5j6moqDAFBQUmKyvLTJkyxXzrW9/qc7mS88Z9boxNpp83PA4YAAALDat7AAAAgDsIAAAAWIgAAACAhQgAAABYiAAAAICFCAAAAFiIAAAAgIUIAAAAWIgAAACAhQgAAABYiAAAAICF/gflNVqdnE5hMwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.random.seed(42)\n",
    "plt.figure(figsize=(6, 3))\n",
    "\n",
    "youtube = np.random.normal(232, 2, 90)\n",
    "vimeo = np.random.normal(219, 2, 90)\n",
    "\n",
    "plt.hist(youtube, edgecolor='k', linewidth=1.2, bins=11, color='r', alpha=0.5, label=\"YouTube\")\n",
    "plt.hist(vimeo, edgecolor='k', linewidth=1.2, bins=11, color='b', alpha=0.5, label=\"Vimeo\")\n",
    "# plt.legend()\n",
    "# plt.xlabel(\"Minutes of viewing\");\n",
    "\n",
    "# plt.savefig(\"two_dists.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Question\n",
    "\n",
    "* What is the mean of the distriution of the difference between the means?\n",
    "\n",
    "```python\n",
    "youtube = np.random.normal(232, 2, 90)\n",
    "vimeo = np.random.normal(219, 2, 90)\n",
    "means_difference.append(np.mean(youtube) - np.mean(vimeo))\n",
    "```\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "means_difference = []\n",
    "\n",
    "for _ in range(1000):\n",
    "    youtube = np.random.normal(232, 2, 90)\n",
    "    vimeo = np.random.normal(219, 2, 90)\n",
    "\n",
    "    means_difference.append(np.mean(youtube) - np.mean(vimeo))\n",
    "    \n",
    "_ = plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Assessing the Variability of Our Statistic: Similar Samples\n",
    "\n",
    "* What does the difference between two sample of the same distributions look like on average?\n",
    "\n",
    "  \n",
    "```python\n",
    "youtube_1 = np.random.normal(232, 2, 90)\n",
    "youtube_2 = np.random.normal(232, 2, 90)\n",
    "means_difference.append(np.mean(youtube_1) - np.mean(youtube_2))\n",
    "```\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "means_difference = []\n",
    "\n",
    "for _ in range(1000):\n",
    "    youtube_1 = np.random.normal(232, 2, 90)\n",
    "    youtube_2 = np.random.normal(232, 2, 90)\n",
    "\n",
    "    means_difference.append(np.mean(youtube_1) - np.mean(youtube_2))\n",
    "    \n",
    "_ = plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12,)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Assessing the Variability of Our Statistic: Similar Samples -- Cont'd\n",
    "\n",
    "* Data in `youtube_1` and `youtube_2` comes from the same distribution\n",
    "* Instead of generating new datasets, we can:\n",
    "  1. Generate the data in the same call to `np.random.normal` and then  split the data into `youtube_1` and `youtube_2`\n",
    "  2. We could even permute the data and split it randomly across `youtube_1` and `youtube_2`\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shuffled data is [5 3 4 1 2]\n",
      "Chunk 1 is [5 3]\n",
      "Chunk 2 is [4 1 2]\n"
     ]
    }
   ],
   "source": [
    "a = np.array([1, 2, 3, 4, 5])\n",
    "\n",
    "# shuffles the data in place\n",
    "np.random.shuffle(a)\n",
    "print(f\"Shuffled data is {a}\")\n",
    "\n",
    "print(f\"Chunk 1 is {a[:2]}\")\n",
    "print(f\"Chunk 2 is {a[2:]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1. Generating the data in the same call, shuffling and splitting the data does not change the results\n",
    "\n",
    "plt.figure(figsize=(12, 6))\n",
    "means_difference = []\n",
    "\n",
    "for _ in range(1000):\n",
    "    # Generate 180 days worth of data in each iteration\n",
    "    all_data = np.random.normal(232, 2, 180)\n",
    "    \n",
    "    # shuffle the data and then split it over youtube_1 and youtube_2\n",
    "    np.random.shuffle(all_data)\n",
    "    youtube_1 = all_data[0:90]\n",
    "    youtube_2 = all_data[90:]\n",
    "    \n",
    "    means_difference.append(np.mean(youtube_1) - np.mean(youtube_2))\n",
    "    \n",
    "_ = plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Question\n",
    "\n",
    "* What would the distribution of differences between the means of concatenated (merged) and permutated (shuffled) Vimeo and YouTube looks like? \n",
    "\n",
    "```\n",
    "all_data = np.concatenate([youtube, vimeo])\n",
    "\n",
    "for _ in range(1000):\n",
    "    np.random.shuffle(all_data)\n",
    "    S1 = all_data[:90]\n",
    "    S2 = all_data[90:]\n",
    "    means_difference.append(np.mean(S1) - np.mean(S2))\n",
    "```\n",
    "\n",
    "<center><img src=\"https://www.dropbox.com/scl/fi/xfsns1fkpbviqqc2yvpnq/unknown_mean.png?rlkey=kidq2va5ipbjztys8z16mpwml&dl=1\" alt=\"drawing\" style=\"width:600px\"></center>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "###  Concatenated and Permutated YouTube and Vimeo Samples \n",
    "\n",
    "<img src=\"https://www.dropbox.com/scl/fi/s4nnf1xbtpe1zkcjcajbg/permutation.png?rlkey=pd0l8g1a8i3f0nvdi5840r1yr&dl=1\" alt=\"drawing\" style=\"width:600px\">\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9oAAAH5CAYAAAB+sEb2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAoP0lEQVR4nO3de5DV9X34/9cGlgUEFnFlL3UhlGDHgqEBEpUYZY0SNwYCGC/RMZAgqVGZMEhi0baSTisZ1BgHBmoaCxqx2EyDMjWJ4gXQMY6gYhBbxCMKFla8wbLELrfz/aM/z8+V6+p7+Rzg8Zj5zOzncs55nZ3PMD79nM/Zknw+nw8AAAAgic9kPQAAAAAcTYQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASap/1AJ/Enj17YuPGjdG1a9coKSnJehwAAACOcvl8PrZt2xY1NTXxmc8c+Jr1ERnaGzdujNra2qzHAAAA4BizYcOGOOmkkw54zBEZ2l27do2I/3uD3bp1y3gaAAAAjnaNjY1RW1tb6NEDOSJD+8OPi3fr1k1oAwAAcNgcyu3LvgwNAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkFD7rAcAAA6fkSNHRi6Xy3qMotW3b99YtGhR1mMAcIQT2gBwDMnlcvHymleitHtN1qMUnZ1bNmY9AgBHCaENAMeY0u41UXPl7KzHKDobf3l11iMAcJRwjzYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQELtW/uAZcuWxS233BLPPfdcbNq0KRYuXBijRo0q7C8pKdnn42bMmBE/+tGPIiJi2LBhsXTp0hb7L7nkkliwYEFrxwGAfRo5cmTkcrmsxyg6uVwuoktl1mMAwFGt1aG9ffv2GDhwYHz3u9+NCy+8cK/9mzZtarH+u9/9LsaPH7/XsRMmTIh/+Id/KKx36tSptaMAwH7lcrl4ec0rUdq9JutRisrO5uYo7ZL1FABwdGt1aNfX10d9ff1+91dVVbVYf/DBB6Ouri7+/M//vMX2zp0773Xs/jQ3N0dzc3NhvbGxsRUTA3CsKu1eEzVXzs56jKLyxq2jsx4BAI56bXqP9ltvvRUPPfRQjB8/fq998+fPj4qKiujfv39MmTIltm3btt/nmT59epSXlxeW2trathwbAAAAPrFWX9Fujbvvvju6du0aY8aMabH98ssvjz59+kRVVVW89NJLMXXq1HjxxRdj8eLF+3yeqVOnxuTJkwvrjY2NYhsAAICi1Kah/a//+q9x+eWXR8eOHVtsnzBhQuHnAQMGRL9+/WLIkCHx/PPPx6BBg/Z6nrKysigrK2vLUQEAACCJNvvo+JNPPhlr1qyJK6+88qDHDho0KEpLS2Pt2rVtNQ4AAAAcFm0W2nfddVcMHjw4Bg4ceNBjV69eHTt37ozq6uq2GgcAAAAOi1Z/dLypqSleffXVwvq6deti5cqV0aNHj+jVq1dE/N891L/+9a/jtttu2+vxuVwu5s+fH1//+tejoqIiXn755bjuuuviC1/4Qnz5y1/+FG8FAAAAstfq0F6xYkXU1dUV1j/8krKxY8fGvHnzIiJiwYIFkc/n49vf/vZej+/QoUM89thjcccdd0RTU1PU1tbGBRdcEDfddFO0a9fuE74NAAAAKA6tDu1hw4ZFPp8/4DHf//734/vf//4+99XW1sbSpUtb+7IAAABwRGjTv6MNAAAAxxqhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhNpnPQAAQDHY1fh25Jreiv79+2c9StHp27dvLFq0KOsxAI4YQhsAICLye3ZF8558rH2rKetRisrOLRuzHgHgiCO0AQD+P6Xda6LmytlZj1FUNv7y6qxHADjiuEcbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJtTq0ly1bFiNGjIiampooKSmJBx54oMX+cePGRUlJSYvl9NNPb3FMc3NzTJw4MSoqKuK4446LkSNHxptvvvmp3ggAAAAUg1aH9vbt22PgwIExa9as/R5z/vnnx6ZNmwrLb3/72xb7J02aFAsXLowFCxbEU089FU1NTfGNb3wjdu/e3fp3AAAAAEWkfWsfUF9fH/X19Qc8pqysLKqqqva5b+vWrXHXXXfFr371qzj33HMjIuLee++N2traePTRR+NrX/vaXo9pbm6O5ubmwnpjY2NrxwYAAIDDok3u0V6yZEn07NkzTj755JgwYUJs3ry5sO+5556LnTt3xvDhwwvbampqYsCAAfH000/v8/mmT58e5eXlhaW2trYtxgYAAIBPLXlo19fXx/z58+Pxxx+P2267LZYvXx7nnHNO4Yp0Q0NDdOjQIY4//vgWj6usrIyGhoZ9PufUqVNj69athWXDhg2pxwYAAIAkWv3R8YO55JJLCj8PGDAghgwZEr17946HHnooxowZs9/H5fP5KCkp2ee+srKyKCsrSz0qAAAAJNfmf96ruro6evfuHWvXro2IiKqqqtixY0e8//77LY7bvHlzVFZWtvU4AAAA0KaSX9H+uHfffTc2bNgQ1dXVERExePDgKC0tjcWLF8fFF18cERGbNm2Kl156KWbMmNHW4wAcVUaOHBm5XC7rMYpSLpeL6OJ/4AIAh1+rQ7upqSleffXVwvq6deti5cqV0aNHj+jRo0dMmzYtLrzwwqiuro7XX389brjhhqioqIjRo0dHRER5eXmMHz8+rrvuujjhhBOiR48eMWXKlDj11FML30IOwKHJ5XLx8ppXorR7TdajFJ2dzc1R2iXrKQCAY1GrQ3vFihVRV1dXWJ88eXJERIwdOzbmzJkTq1atinvuuSe2bNkS1dXVUVdXF/fff3907dq18Jjbb7892rdvHxdffHF88MEH8dWvfjXmzZsX7dq1S/CWAI4tpd1roubK2VmPUXTeuHV01iMAAMeoVof2sGHDIp/P73f/ww8/fNDn6NixY8ycOTNmzpzZ2pcHAACAotbmX4YGAAAAxxKhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJNTq0F62bFmMGDEiampqoqSkJB544IHCvp07d8b1118fp556ahx33HFRU1MT3/nOd2Ljxo0tnmPYsGFRUlLSYrn00ks/9ZsBAACArLU6tLdv3x4DBw6MWbNm7bXvT3/6Uzz//PPxd3/3d/H888/Hb37zm3jllVdi5MiRex07YcKE2LRpU2G58847P9k7AAAAgCLSvrUPqK+vj/r6+n3uKy8vj8WLF7fYNnPmzPjSl74U69evj169ehW2d+7cOaqqqlr78gAAAFDU2vwe7a1bt0ZJSUl07969xfb58+dHRUVF9O/fP6ZMmRLbtm3b73M0NzdHY2NjiwUAAACKUauvaLfG//7v/8bf/M3fxGWXXRbdunUrbL/88sujT58+UVVVFS+99FJMnTo1Xnzxxb2uhn9o+vTp8ZOf/KQtRwUAAIAk2iy0d+7cGZdeemns2bMnZs+e3WLfhAkTCj8PGDAg+vXrF0OGDInnn38+Bg0atNdzTZ06NSZPnlxYb2xsjNra2rYaHQAAAD6xNgntnTt3xsUXXxzr1q2Lxx9/vMXV7H0ZNGhQlJaWxtq1a/cZ2mVlZVFWVtYWowIAAEBSyUP7w8heu3ZtPPHEE3HCCScc9DGrV6+OnTt3RnV1depxAAAA4LBqdWg3NTXFq6++Wlhft25drFy5Mnr06BE1NTXxrW99K55//vn4z//8z9i9e3c0NDRERESPHj2iQ4cOkcvlYv78+fH1r389Kioq4uWXX47rrrsuvvCFL8SXv/zldO8MAAAAMtDq0F6xYkXU1dUV1j+8d3rs2LExbdq0WLRoUURE/NVf/VWLxz3xxBMxbNiw6NChQzz22GNxxx13RFNTU9TW1sYFF1wQN910U7Rr1+5TvBUAAADIXqtDe9iwYZHP5/e7/0D7IiJqa2tj6dKlrX1ZAAAAOCK0+d/RBgAAgGOJ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACbXPegAAAIrXrsa3I9f0VvTv3z/rUYpS3759Y9GiRVmPARQZoQ0AwH7l9+yK5j35WPtWU9ajFJ2dWzZmPQJQpIQ2AAAHVNq9JmqunJ31GEVn4y+vznoEoEi5RxsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCrQ7tZcuWxYgRI6KmpiZKSkrigQceaLE/n8/HtGnToqamJjp16hTDhg2L1atXtzimubk5Jk6cGBUVFXHcccfFyJEj48033/xUbwQAAACKQatDe/v27TFw4MCYNWvWPvfPmDEjfvazn8WsWbNi+fLlUVVVFeedd15s27atcMykSZNi4cKFsWDBgnjqqaeiqakpvvGNb8Tu3bs/+TsBAACAItC+tQ+or6+P+vr6fe7L5/Px85//PG688cYYM2ZMRETcfffdUVlZGffdd1/89V//dWzdujXuuuuu+NWvfhXnnntuRETce++9UVtbG48++mh87Wtf+xRvBwAAALKV9B7tdevWRUNDQwwfPrywraysLM4+++x4+umnIyLiueeei507d7Y4pqamJgYMGFA45uOam5ujsbGxxQIAAADFKGloNzQ0REREZWVli+2VlZWFfQ0NDdGhQ4c4/vjj93vMx02fPj3Ky8sLS21tbcqxAQAAIJk2+dbxkpKSFuv5fH6vbR93oGOmTp0aW7duLSwbNmxINisAAACklDS0q6qqIiL2ujK9efPmwlXuqqqq2LFjR7z//vv7PebjysrKolu3bi0WAAAAKEZJQ7tPnz5RVVUVixcvLmzbsWNHLF26NIYOHRoREYMHD47S0tIWx2zatCleeumlwjEAAABwpGr1t443NTXFq6++Wlhft25drFy5Mnr06BG9evWKSZMmxc033xz9+vWLfv36xc033xydO3eOyy67LCIiysvLY/z48XHdddfFCSecED169IgpU6bEqaeeWvgWcgAAADhStTq0V6xYEXV1dYX1yZMnR0TE2LFjY968efHjH/84Pvjgg7j66qvj/fffj9NOOy0eeeSR6Nq1a+Ext99+e7Rv3z4uvvji+OCDD+KrX/1qzJs3L9q1a5fgLQEAAEB2Wh3aw4YNi3w+v9/9JSUlMW3atJg2bdp+j+nYsWPMnDkzZs6c2dqXBwAAgKLWJt86DgAAAMcqoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQu2zHgDgUIwcOTJyuVzWYxSdXC4X0aUy6zEAAPgIoQ0cEXK5XLy85pUo7V6T9ShFZWdzc5R2yXoKAAA+SmgDR4zS7jVRc+XsrMcoKm/cOjrrEQAA+Bj3aAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCyUP7s5/9bJSUlOy1XHPNNRERMW7cuL32nX766anHAAAAgEy0T/2Ey5cvj927dxfWX3rppTjvvPPioosuKmw7//zzY+7cuYX1Dh06pB4DAAAAMpE8tE888cQW6z/96U+jb9++cfbZZxe2lZWVRVVVVeqXBgAAgMy16T3aO3bsiHvvvTe+973vRUlJSWH7kiVLomfPnnHyySfHhAkTYvPmzQd8nubm5mhsbGyxAAAAQDFq09B+4IEHYsuWLTFu3LjCtvr6+pg/f348/vjjcdttt8Xy5cvjnHPOiebm5v0+z/Tp06O8vLyw1NbWtuXYAAAA8Ikl/+j4R911111RX18fNTU1hW2XXHJJ4ecBAwbEkCFDonfv3vHQQw/FmDFj9vk8U6dOjcmTJxfWGxsbxTYAAABFqc1C+4033ohHH300fvOb3xzwuOrq6ujdu3esXbt2v8eUlZVFWVlZ6hEBAAAguTb76PjcuXOjZ8+eccEFFxzwuHfffTc2bNgQ1dXVbTUKAAAAHDZtEtp79uyJuXPnxtixY6N9+///onlTU1NMmTIl/vCHP8Trr78eS5YsiREjRkRFRUWMHj26LUYBAACAw6pNPjr+6KOPxvr16+N73/tei+3t2rWLVatWxT333BNbtmyJ6urqqKuri/vvvz+6du3aFqMAAADAYdUmoT18+PDI5/N7be/UqVM8/PDDbfGSAAAAUBTa9M97AQAAwLFGaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAklD+1p06ZFSUlJi6WqqqqwP5/Px7Rp06KmpiY6deoUw4YNi9WrV6ceAwAAADLRJle0+/fvH5s2bSosq1atKuybMWNG/OxnP4tZs2bF8uXLo6qqKs4777zYtm1bW4wCAAAAh1WbhHb79u2jqqqqsJx44okR8X9Xs3/+85/HjTfeGGPGjIkBAwbE3XffHX/605/ivvvua4tRAAAA4LBqk9Beu3Zt1NTURJ8+feLSSy+N1157LSIi1q1bFw0NDTF8+PDCsWVlZXH22WfH008/vd/na25ujsbGxhYLAAAAFKPkoX3aaafFPffcEw8//HD8y7/8SzQ0NMTQoUPj3XffjYaGhoiIqKysbPGYysrKwr59mT59epSXlxeW2tra1GMDAABAEslDu76+Pi688MI49dRT49xzz42HHnooIiLuvvvuwjElJSUtHpPP5/fa9lFTp06NrVu3FpYNGzakHhsAAACSaPM/73XcccfFqaeeGmvXri18+/jHr15v3rx5r6vcH1VWVhbdunVrsQAAAEAxavPQbm5ujv/6r/+K6urq6NOnT1RVVcXixYsL+3fs2BFLly6NoUOHtvUoAAAA0Obap37CKVOmxIgRI6JXr16xefPm+Md//MdobGyMsWPHRklJSUyaNCluvvnm6NevX/Tr1y9uvvnm6Ny5c1x22WWpRwEAAIDDLnlov/nmm/Htb3873nnnnTjxxBPj9NNPj2eeeSZ69+4dERE//vGP44MPPoirr7463n///TjttNPikUceia5du6YeBQAAAA675KG9YMGCA+4vKSmJadOmxbRp01K/NAAAAGSuze/RBgAAgGOJ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJBQ+6wHAACAI9Guxrcj1/RW9O/fP+tRik7fvn1j0aJFWY8BmRHaAADwCeT37IrmPflY+1ZT1qMUlZ1bNmY9AmROaAMAwCdU2r0maq6cnfUYRWXjL6/OegTInHu0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBCQhsAAAASEtoAAACQkD/vBUVm5MiRkcvlsh6j6ORyuYgulVmPAQAAByW0ocjkcrl4ec0rUdq9JutRisrO5uYo7ZL1FAAAcHBCG4pQafeaqLlydtZjFJU3bh2d9QgAAHBI3KMNAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAkJLQBAAAgIaENAAAACQltAAAASEhoAwAAQELJQ3v69OnxxS9+Mbp27Ro9e/aMUaNGxZo1a1ocM27cuCgpKWmxnH766alHAQAAgMMueWgvXbo0rrnmmnjmmWdi8eLFsWvXrhg+fHhs3769xXHnn39+bNq0qbD89re/TT0KAAAAHHbtUz/h73//+xbrc+fOjZ49e8Zzzz0XZ511VmF7WVlZVFVVpX55AAAAyFSb36O9devWiIjo0aNHi+1LliyJnj17xsknnxwTJkyIzZs37/c5mpubo7GxscUCAAAAxahNQzufz8fkyZPjzDPPjAEDBhS219fXx/z58+Pxxx+P2267LZYvXx7nnHNONDc37/N5pk+fHuXl5YWltra2LccGAACATyz5R8c/6tprr40//vGP8dRTT7XYfskllxR+HjBgQAwZMiR69+4dDz30UIwZM2av55k6dWpMnjy5sN7Y2Ci2AQAAKEptFtoTJ06MRYsWxbJly+Kkk0464LHV1dXRu3fvWLt27T73l5WVRVlZWVuMCQAAAEklD+18Ph8TJ06MhQsXxpIlS6JPnz4Hfcy7774bGzZsiOrq6tTjAAAAwGGV/B7ta665Ju6999647777omvXrtHQ0BANDQ3xwQcfREREU1NTTJkyJf7whz/E66+/HkuWLIkRI0ZERUVFjB49OvU4AAAAcFglv6I9Z86ciIgYNmxYi+1z586NcePGRbt27WLVqlVxzz33xJYtW6K6ujrq6uri/vvvj65du6YeBwAAAA6rNvno+IF06tQpHn744dQvCwAAFIFdjW9Hrumt6N+/f9ajFKW+ffvGokWLsh6DNtam3zoOAAAcW/J7dkXznnysfasp61GKzs4tG7MegcNEaAMAAEmVdq+JmitnZz1G0dn4y6uzHoHDJPmXoQEAAMCxTGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgITaZz0AAADAsWBX49uRa3or+vfvn/UoRadv376xaNGirMdIRmgDAAAcBvk9u6J5Tz7WvtWU9ShFZeeWjVmPkJzQBgAAOExKu9dEzZWzsx6jqGz85dVZj5Cce7QBAAAgIaENAAAACfnoOJkYOXJk5HK5rMcoSrlcLqJLZdZjAAAAn5DQJhO5XC5eXvNKlHavyXqUorOzuTlKu2Q9BQAA8EkJbTLjiyD27Y1bR2c9AgAA8Cm4RxsAAAASEtoAAACQkNAGAACAhIQ2AAAAJCS0AQAAICGhDQAAAAkJbQAAAEhIaAMAAEBC7bMe4Gg3cuTIyOVyWY9RdHK5XESXyqzHAAAASE5ot7FcLhcvr3klSrvXZD1KUdnZ3BylXbKeAgAAID2hfRiUdq+JmitnZz1GUXnj1tFZjwAAANAm3KMNAAAACQltAAAASEhoAwAAQEJCGwAAABIS2gAAAJCQ0AYAAICEhDYAAAAklGloz549O/r06RMdO3aMwYMHx5NPPpnlOAAAAPCpZRba999/f0yaNCluvPHGeOGFF+IrX/lK1NfXx/r167MaCQAAAD61knw+n8/ihU877bQYNGhQzJkzp7DtlFNOiVGjRsX06dNbHNvc3BzNzc2F9a1bt0avXr1iw4YN0a1bt8M28yfxpS99KdasfTXal1dnPUpR2fX+mxGfaef3sg9+N/vm97Jvfi/753ezb34v++d3s29+L/vnd7Nvfi/753ezb7u2boq/6Pe5ePbZZ7Me5YAaGxujtrY2tmzZEuXl5Qc8NpPQ3rFjR3Tu3Dl+/etfx+jRowvbf/jDH8bKlStj6dKlLY6fNm1a/OQnPzncYwIAAEALGzZsiJNOOumAx7Q/TLO08M4778Tu3bujsrKyxfbKyspoaGjY6/ipU6fG5MmTC+t79uyJ9957L0444YQoKSlp83mL1Yf/R+VIuLJPdpwnHCrnCofCecKhcq5wKJwnHIpiOU/y+Xxs27YtampqDnpsJqH9oY9Hcj6f32c4l5WVRVlZWYtt3bt3b8vRjijdunXzDxMH5TzhUDlXOBTOEw6Vc4VD4TzhUBTDeXKwj4x/KJMvQ6uoqIh27drtdfV68+bNe13lBgAAgCNJJqHdoUOHGDx4cCxevLjF9sWLF8fQoUOzGAkAAACSyOyj45MnT44rrrgihgwZEmeccUb84he/iPXr18dVV12V1UhHnLKysrjpppv2+lg9fJTzhEPlXOFQOE84VM4VDoXzhENxJJ4nmf15r4iI2bNnx4wZM2LTpk0xYMCAuP322+Oss87KahwAAAD41DINbQAAADjaZHKPNgAAABythDYAAAAkJLQBAAAgIaENAAAACQnto8TIkSOjV69e0bFjx6iuro4rrrgiNm7cmPVYFJnXX389xo8fH3369IlOnTpF375946abboodO3ZkPRpF5p/+6Z9i6NCh0blz5+jevXvW41BEZs+eHX369ImOHTvG4MGD48knn8x6JIrMsmXLYsSIEVFTUxMlJSXxwAMPZD0SRWj69OnxxS9+Mbp27Ro9e/aMUaNGxZo1a7IeiyIzZ86c+PznPx/dunWLbt26xRlnnBG/+93vsh7rkAjto0RdXV38+7//e6xZsyb+4z/+I3K5XHzrW9/KeiyKzH//93/Hnj174s4774zVq1fH7bffHv/8z/8cN9xwQ9ajUWR27NgRF110UfzgBz/IehSKyP333x+TJk2KG2+8MV544YX4yle+EvX19bF+/fqsR6OIbN++PQYOHBizZs3KehSK2NKlS+Oaa66JZ555JhYvXhy7du2K4cOHx/bt27MejSJy0kknxU9/+tNYsWJFrFixIs4555z45je/GatXr856tIPy572OUosWLYpRo0ZFc3NzlJaWZj0OReyWW26JOXPmxGuvvZb1KBShefPmxaRJk2LLli1Zj0IROO2002LQoEExZ86cwrZTTjklRo0aFdOnT89wMopVSUlJLFy4MEaNGpX1KBS5t99+O3r27BlLly6Ns846K+txKGI9evSIW265JcaPH5/1KAfkivZR6L333ov58+fH0KFDRTYHtXXr1ujRo0fWYwBFbseOHfHcc8/F8OHDW2wfPnx4PP300xlNBRwttm7dGhHhv0nYr927d8eCBQti+/btccYZZ2Q9zkEJ7aPI9ddfH8cdd1yccMIJsX79+njwwQezHokil8vlYubMmXHVVVdlPQpQ5N55553YvXt3VFZWttheWVkZDQ0NGU0FHA3y+XxMnjw5zjzzzBgwYEDW41BkVq1aFV26dImysrK46qqrYuHChfGXf/mXWY91UEK7iE2bNi1KSkoOuKxYsaJw/I9+9KN44YUX4pFHHol27drFd77znXBnwLGhtedKRMTGjRvj/PPPj4suuiiuvPLKjCbncPok5wl8XElJSYv1fD6/1zaA1rj22mvjj3/8Y/zbv/1b1qNQhP7iL/4iVq5cGc8880z84Ac/iLFjx8bLL7+c9VgH1T7rAdi/a6+9Ni699NIDHvPZz3628HNFRUVUVFTEySefHKecckrU1tbGM888c0R8tIJPp7XnysaNG6Ouri7OOOOM+MUvftHG01EsWnuewEdVVFREu3bt9rp6vXnz5r2ucgMcqokTJ8aiRYti2bJlcdJJJ2U9DkWoQ4cO8bnPfS4iIoYMGRLLly+PO+64I+68886MJzswoV3EPgznT+LDK9nNzc0pR6JIteZc+Z//+Z+oq6uLwYMHx9y5c+Mzn/HBlmPFp/k3BTp06BCDBw+OxYsXx+jRowvbFy9eHN/85jcznAw4EuXz+Zg4cWIsXLgwlixZEn369Ml6JI4Q+Xz+iGgcoX0UePbZZ+PZZ5+NM888M44//vh47bXX4u///u+jb9++rmbTwsaNG2PYsGHRq1evuPXWW+Ptt98u7KuqqspwMorN+vXr47333ov169fH7t27Y+XKlRER8bnPfS66dOmS7XBkZvLkyXHFFVfEkCFDCp+IWb9+ve95oIWmpqZ49dVXC+vr1q2LlStXRo8ePaJXr14ZTkYxueaaa+K+++6LBx98MLp27Vr4tEx5eXl06tQp4+koFjfccEPU19dHbW1tbNu2LRYsWBBLliyJ3//+91mPdlD+vNdRYNWqVfHDH/4wXnzxxdi+fXtUV1fH+eefH3/7t38bf/Znf5b1eBSRefPmxXe/+9197vNPAR81bty4uPvuu/fa/sQTT8SwYcMO/0AUjdmzZ8eMGTNi06ZNMWDAgLj99tv9KR5aWLJkSdTV1e21fezYsTFv3rzDPxBFaX/f7TB37twYN27c4R2GojV+/Ph47LHHYtOmTVFeXh6f//zn4/rrr4/zzjsv69EOSmgDAABAQm7OBAAAgISENgAAACQktAEAACAhoQ0AAAAJCW0AAABISGgDAABAQkIbAAAAEhLaAAAAkJDQBgAAgISENgAAACQktAEAACCh/wdFeRAZS0CvRgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "means_difference = []\n",
    "\n",
    "youtube = np.random.normal(232, 2, 90)\n",
    "vimeo = np.random.normal(219, 2, 90)\n",
    "all_data = np.concatenate([youtube, vimeo])\n",
    "\n",
    "for _ in range(1000):\n",
    "    np.random.shuffle(all_data)\n",
    "    youtube_s = all_data[:90]\n",
    "    vimeo_s = all_data[90:]\n",
    "    means_difference.append(np.mean(youtube_s) - np.mean(vimeo_s))\n",
    "    \n",
    "_ = plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Question 1\n",
    "\n",
    "* What would the distribution of differences between the means of merged and permutated Vimeo and YouTube look like? \n",
    "<br>\n",
    "<br>\n",
    "\n",
    "#### __Answer:__\n",
    "\n",
    "* The mean of the distribution of differences is 0 since the experiment is equivalent to sampling from a single large population of both YouTube and Vimeo viewers.\n",
    "<br><br>\n",
    "\n",
    "### Question 2\n",
    "\n",
    "* What would the distribution of the shuffled difference between Vimeo and YouTube if the samples are not balanced?\n",
    " * Say, for instance, one sample has 90 and the second has four times more individuals.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "means_difference = []\n",
    "\n",
    "youtube = np.random.normal(232, 2, 90)\n",
    "vimeo = np.random.normal(219, 2, 360)\n",
    "all_data = np.concatenate([youtube, vimeo])\n",
    "\n",
    "for _ in range(1000):\n",
    "    np.random.shuffle(all_data)\n",
    "    youtube_s = all_data[:90]\n",
    "    vimeo_s = all_data[90:]\n",
    "    means_difference.append(np.mean(youtube_s) - np.mean(vimeo_s))\n",
    "    \n",
    "_ = plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Original Observation Vs. Distribution Under the Null Hypothesis\n",
    "\n",
    "\n",
    "<img src=\"https://www.dropbox.com/scl/fi/wf5t4bftvg3ynaxnwgksw/youtube_vimeo_difference.png?rlkey=n6fku7w4wtjjjausa9f6er3gp&dl=1\" alt=\"drawing\" style=\"width:1200px\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "means_difference = []\n",
    "\n",
    "youtube = np.random.normal(232, 2, 90)\n",
    "vimeo = np.random.normal(219, 2, 360)\n",
    "all_data = np.concatenate([youtube, vimeo])\n",
    "\n",
    "for _ in range(1000):\n",
    "    np.random.shuffle(all_data)\n",
    "    youtube_s = all_data[:90]\n",
    "    vimeo_s = all_data[90:]\n",
    "    means_difference.append(np.mean(youtube_s) - np.mean(vimeo_s))\n",
    "    \n",
    "plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)\n",
    "plt.scatter(np.mean(youtube) - np.mean(vimeo), 2, color=\"r\", s=40)\n",
    "plt.ylim(-10, 230);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### What Have We Learned?\n",
    "\n",
    "* The difference between the means of samples from the same distribution has a mean close to 0\n",
    "\n",
    "* This is very useful for testing the null hypothesis\n",
    "  * Under the null hypothesis, we assume that the difference between the values observed across two samples is attributed to chance alone \n",
    "\n",
    "* Comparing the observed difference between the means of two samples to that obtained if samples are from the same distribution shows how similar/dissimilar the samples while accounting for sampling variability\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Comparing Samples from Similar Distributions\n",
    "\n",
    "* What does a plausible observed difference of means look like?\n",
    "\n",
    "* Ex. Is there a difference between the spending of males and females on your e-commerce platform?\n",
    "\n",
    "* Again we will generate the data from two distributions \n",
    " * The distributions are different but close ($\\mu_{male} = 466$ and $\\mu_{female} = 450$)\n",
    " * The standard deviation is the same for both populations but relatively large ($\\sigma = 89$)  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Difference between means is 8.356736308940526\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(129)\n",
    "\n",
    "male_spending = np.random.normal(466, 89, 230)\n",
    "female_spending = np.random.normal(450, 89, 190)\n",
    "\n",
    "diff_of_means = np.mean(male_spending) - np.mean(female_spending)\n",
    "print(f\"Difference between means is {diff_of_means}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_data = np.concatenate([male_spending, female_spending])\n",
    "means_difference = []\n",
    "\n",
    "for _ in range(1000):\n",
    "    np.random.shuffle(all_data)\n",
    "    male_s = all_data[:230]\n",
    "    female_s = all_data[230:]\n",
    "    means_difference.append(np.mean(male_s) - np.mean(female_s))\n",
    "\n",
    "plt.figure(figsize=(6, 3))\n",
    "plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)\n",
    "plt.scatter(np.mean(male_spending) - np.mean(female_spending), 0, color=\"r\", s=40);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### How Surprising Is an Observed Statistic?\n",
    "\n",
    "* Out of the 1000 permutations generated, how many were equal to or larger than the observed statistic?\n",
    "  * We can count the number of simulated statistics that were equal to or larger than the overserved statistic.\n",
    "\n",
    "* The larger that number, the less \"surprised\" we are by the observed statistic.\n",
    "\n",
    "* The smaller that numer, the less likely that the spending of males and females are from the same distribution.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "179"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(np.array(means_difference) >= diff_of_means)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### The P-Value\n",
    "\n",
    "The p-value is a convention used to estimate how far out into the tails is considered \"surprising\"\n",
    "  * Based on the area in the tail, either upper, lower or both\n",
    "     * Since we are working on the empirical distribution, it's easy to compute"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### The P-Value -- Cont'd\n",
    "\n",
    "The p-value is the chance, based on the model in the null hypothesis, that the test statistic is equal to the value that was observed in the data or is even further in the direction of the alternative.\n",
    "\n",
    "* If a p-value is small, then the observed statistic is far away from what the null hypothesis predicts.\n",
    "  * The null hypothesis is unlikely and the data supports the alternative.\n",
    "  * Typically, if p-value is less than 5%, it's considered “statistically significant”\n",
    "    * Reasonable to consider that occurrence in a frequency of 1/20 to be surprising.\n",
    "\n",
    "* The significance level ($\\alpha$) is the probability of rejecting the null hypothesis when it is true. (Type-I error rate)\n",
    "  * A significance level of 0.05 indicates a 5% risk of concluding that a difference exists when there is no actual difference."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Hypothesis Testing Approach (General Algorithm)\n",
    "\n",
    "Strategy used for testing hypotheses will consist of the following steps:\n",
    "\n",
    "1. We will assume that the null hypothesis is true and will compute some test statistic between the samples, or between the sample and population parameter\n",
    "  * Identifying the appropriate test statistic is critical \n",
    "    * This is the observed statistic under the null hypothesis\n",
    "\n",
    "2. Use permutations or bootstrapping to build a distribution of the test statistic under the null hypothesis\n",
    "  \n",
    "3. Use the inferred distribution of the test statistic to infer whether the observed statistic is plausible under the null hypothesis\n",
    "  * How likely are we to observe a value as extreme or more extreme than the observed statistic?\n",
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Comparing a Sample to a Distribution Parameter\n",
    "\n",
    "* Ex. The national average for the math subscore of the SAT is 525, with a standard deviation of 20. Does a sample of 100 students who followed your curriculum far better than the national average?\n",
    "\n",
    "  1. We can generate bootstrap samples and bootstrap statistics (replicates) and infer the 95% confidence interval for the national SAT scores\n",
    "  2. If the sample mean of the students who followed your curriculum is included in the 95% confidence interval, then we don't have enough evidence to discard the null hypothesis\n",
    "    * I.e., the scores obtained by our 100 students are from the theoretical distribution with $\\mu = 525$ and $\\sigma = 20$\n",
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "\n",
    "<img src=\"https://www.dropbox.com/s/997cfpe674ujdub/dist_mean.png?dl=1\">  \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Example\n",
    "\n",
    "* Suppose we have two samples that represent Biki ride durations for two distinct neighborhoods `a` and `b`\n",
    "* Instead of working with actual data, sample the data for the two neighborhoods from a Gaussian with a mean ($\\mu=20$) and a standard deviation ($\\sigma = 4$)\n",
    "    * We will sample 15 bike rides for each of regions `a` and `b`\n",
    "    * In our scenario, this amounts to measuring the duration of 30 bike rides in neighborhoods `a` and `b` combined.\n",
    "<br><br>\n",
    "1. Compute and compare the means of datasets `a` and `b`\n",
    "  * Based on the observed difference between the means, do you believe there is a statistically significant difference between samples `a` and `b`?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(22.748033269279322, 20.794818308826077)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.seed(0)\n",
    "a = np.random.normal(20, 4, 15)\n",
    "b = np.random.normal(20, 4, 15)\n",
    "np.mean(a), np.mean(b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "### Example -- Cont'd\n",
    "\n",
    "2. Compute and compare the means of datasets `a` and `b`\n",
    "  * Based on the observed difference between the means, do you believe there is a statistically significant difference between samples `a` and `b`?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Example -- Cont'd\n",
    "\n",
    "2.$~$Draw the two KDEs representing the datasets `a` and `b`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy.stats as stats\n",
    "\n",
    "x_axis = np.arange(0, 40, 0.5)\n",
    "kde_1 = sp.stats.gaussian_kde(a, bw_method=1)\n",
    "kde_2 = sp.stats.gaussian_kde(b, bw_method=1)\n",
    "x_densities_1 = kde_1.evaluate(x_axis)\n",
    "x_densities_2 = kde_2.evaluate(x_axis)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 4))\n",
    "plt.plot(x_axis, x_densities_1, label=\"dataset a\")\n",
    "plt.plot(x_axis, x_densities_2, label=\"dataset b\")\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Example -- Cont'd\n",
    "\n",
    "3.$~$Visually, does dataset `a` differ significantly from dataset `b`?\n",
    "\n",
    "4.$~$Use the resampling method covered in class to test whether or not there is a statistically significant difference between datasets `a` and `b`\n",
    " * Plot a histogram of the means obtained during resampling and the observed value\n",
    " * Provide a p-value to support your claim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "### Example -- Cont'd\n",
    "\n",
    "5. Use the resampling method covered in class to test whether or not there is a statistically significant difference between datasets `a` and `b`\n",
    " * Plot a histogram of the means obtained during resampling and the observed value\n",
    " * Provide a p-value to support your claim."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 4))\n",
    "\n",
    "all_data = np.concatenate([a, b])\n",
    "means_difference = []\n",
    "\n",
    "for _ in range(1000):\n",
    "    np.random.shuffle(all_data)\n",
    "    a_subset = all_data[:15]\n",
    "    b_subset = all_data[15:]\n",
    "    means_difference.append(np.mean(a_subset) - np.mean(b_subset))\n",
    "    \n",
    "plt.hist(means_difference, edgecolor='k', linewidth=1.2, bins=12)\n",
    "plt.scatter(np.mean(a) - np.mean(b), 0, color=\"r\", s=40);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "118"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(means_difference >= (np.mean(a) - np.mean(b)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.118"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(means_difference >= (np.mean(a) - np.mean(b))) / len(means_difference) "
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "colab": {
   "collapsed_sections": [],
   "name": "PP_1_Introduction.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}