day-3-data-mining/00 - Statistical analysis with Python.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:48:59.227432Z",
     "start_time": "2019-07-02T14:48:59.213333Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>\n",
       "    @font-face {\n",
       "        font-family: \"Computer Modern\";\n",
       "        src: url('http://9dbb143991406a7c655e-aa5fcb0a5a4ec34cff238a2d56ca4144.r56.cf5.rackcdn.com/cmunss.otf');\n",
       "    }\n",
       "    @font-face {\n",
       "        font-family: \"Computer Modern\";\n",
       "        font-weight: bold;\n",
       "        src: url('http://9dbb143991406a7c655e-aa5fcb0a5a4ec34cff238a2d56ca4144.r56.cf5.rackcdn.com/cmunsx.otf');\n",
       "    }\n",
       "    @font-face {\n",
       "        font-family: \"Computer Modern\";\n",
       "        font-style: oblique;\n",
       "        src: url('http://9dbb143991406a7c655e-aa5fcb0a5a4ec34cff238a2d56ca4144.r56.cf5.rackcdn.com/cmunsi.otf');\n",
       "    }\n",
       "    @font-face {\n",
       "        font-family: \"Computer Modern\";\n",
       "        font-weight: bold;\n",
       "        font-style: oblique;\n",
       "        src: url('http://9dbb143991406a7c655e-aa5fcb0a5a4ec34cff238a2d56ca4144.r56.cf5.rackcdn.com/cmunso.otf');\n",
       "    }\n",
       "    h1 {\n",
       "        font-family: Helvetica, serif;\n",
       "    }\n",
       "    h4{\n",
       "        margin-top:12px;\n",
       "        margin-bottom: 3px;\n",
       "       }\n",
       "    div.text_cell_render{\n",
       "        font-family: Computer Modern, \"Helvetica Neue\", Arial, Helvetica, Geneva, sans-serif;\n",
       "        line-height: 145%;\n",
       "        font-size: 130%;\n",
       "        margin-left:auto;\n",
       "        margin-right:auto;\n",
       "    }\n",
       "    .CodeMirror{\n",
       "            font-family: \"Source Code Pro\", source-code-pro,Consolas, monospace;\n",
       "    }\n",
       "    .text_cell_render h5 {\n",
       "        font-weight: 300;\n",
       "        font-size: 22pt;\n",
       "        color: #4057A1;\n",
       "        font-style: italic;\n",
       "        margin-bottom: .5em;\n",
       "        margin-top: 0.5em;\n",
       "        display: block;\n",
       "    }\n",
       "    \n",
       "    .warning{\n",
       "        color: rgb( 240, 20, 20 )\n",
       "        }  \n",
       "</style>\n",
       "<script>\n",
       "    MathJax.Hub.Config({\n",
       "                        TeX: {\n",
       "                           extensions: [\"AMSmath.js\"]\n",
       "                           },\n",
       "                tex2jax: {\n",
       "                    inlineMath: [ ['$','$'], [\"\\\\(\",\"\\\\)\"] ],\n",
       "                    displayMath: [ ['$$','$$'], [\"\\\\[\",\"\\\\]\"] ]\n",
       "                },\n",
       "                displayAlign: 'center', // Change this to 'center' to center equations.\n",
       "                \"HTML-CSS\": {\n",
       "                    styles: {'.MathJax_Display': {\"margin\": 4}}\n",
       "                }\n",
       "        });\n",
       "</script>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.core.display import HTML\n",
    "def css_styling():\n",
    "    styles = open(\"../Data/www/styles/custom.css\", \"r\").read()\n",
    "    return HTML(styles)\n",
    "css_styling()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:49:03.551811Z",
     "start_time": "2019-07-02T14:49:01.768930Z"
    }
   },
   "outputs": [],
   "source": [
    "from copy import copy\n",
    "import json\n",
    "import math\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from random import choice, sample, shuffle\n",
    "import scipy.stats as stat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:49:06.831289Z",
     "start_time": "2019-07-02T14:49:06.822427Z"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:49:23.645076Z",
     "start_time": "2019-07-02T14:49:23.634733Z"
    }
   },
   "outputs": [],
   "source": [
    "def half_frame(sub, xaxis_label, yaxis_label, font_size = 15, padding = -0.02):\n",
    "    \"\"\"Formats frame, axes, and ticks for matplotlib made graphic with half frame.\"\"\"\n",
    "\n",
    "    # Format graph frame and tick marks\n",
    "    sub.yaxis.set_ticks_position('left')\n",
    "    sub.xaxis.set_ticks_position('bottom')\n",
    "    sub.tick_params(axis = 'both', which = 'major', length = 7, width = 2, direction = 'out', pad = 10,\n",
    "                    labelsize = font_size)\n",
    "    sub.tick_params(axis = 'both', which = 'minor', length = 5, width = 2, direction = 'out', labelsize = 10)\n",
    "    for axis in ['bottom','left']:\n",
    "        sub.spines[axis].set_linewidth(2)\n",
    "        sub.spines[axis].set_position((\"axes\", padding))\n",
    "    for axis in ['top','right']:\n",
    "        sub.spines[axis].set_visible(False)\n",
    "\n",
    "    # Format axes\n",
    "    sub.set_xlabel(xaxis_label, fontsize = 1.6 * font_size)\n",
    "    sub.set_ylabel(yaxis_label, fontsize = 1.6 * font_size)\n",
    "    \n",
    "    return\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Synopsis\n",
    "\n",
    "In this unit we will:\n",
    "\n",
    "* Define descriptive and inferential statistics\n",
    "* Demonstrate decriptive and inferential statistics in Python\n",
    "* Cover basic bootstrapping methods for data analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Statistics\n",
    "\n",
    "**Statistics is the science concerned with the study of the collection, analysis, interpretation, presentation, and organization of data.**\n",
    "\n",
    "For concreteness let us consider two examples that hopefully you can relate to.\n",
    "\n",
    "The US Census Bureau collects data on people residing in the US. The data includes measures such as the number of household occupants, their gender, their age, their familial relationships, their incomes, their education levels, and so on. The process for the data collection is called a **census** because the process aims to measure all individuals.\n",
    "\n",
    "Colleges collect applications for their programs. The data collected includes standardized test scores, GPAs, essays, recommendation letters, and so on. No college is able to collect a census, so each must make decisions based on the **sample** of applicants. Each college is able to analyze a sample comprising a very small fraction of all students applying to college each year.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "+++"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to learn more about these situations, let us consider the data we obtained when analyzing the sentiment score of the lines of Othelo and Iago.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:49:34.074023Z",
     "start_time": "2019-07-02T14:49:34.063426Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Othello's data has 470 data points. \n",
      "Iago's data has 537 data points.\n",
      "\n",
      "[1, 1, -3, 3, 1, 2, -3, 2, 3, 2]\n"
     ]
    }
   ],
   "source": [
    "with open('../Data/Day7-Structured-Data-Analysis//iago.json', 'r') as file1:\n",
    "    iago_sentiment = json.load(file1)\n",
    "    \n",
    "\n",
    "with open('../Data/Day7-Structured-Data-Analysis/othello.json', 'r') as file1:\n",
    "    othello_sentiment = json.load(file1)\n",
    "    \n",
    "N_othello = len(othello_sentiment)\n",
    "N_iago = len(iago_sentiment)\n",
    "\n",
    "print(\"Othello's data has {0} data points. \\nIago's data has {1} data points.\\n\".format(N_othello, N_iago) )\n",
    "\n",
    "print(iago_sentiment[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Descriptive statistics\n",
    "\n",
    "The first step in the analysis of data is to to obtain a description that summarizes its statistical properties. There are a number of statistics (that is, measures that can be calculated for that data) that are particularly useful.\n",
    "\n",
    "**Number of observations** is the number of data in the data set. In this case, the number of applicants for which we have GPA values.\n",
    "\n",
    "**Minimum** is the smallest value in the data set.\n",
    "\n",
    "**Maximum** is the largest value in the data set. For the GPA data, this is presumably 4.0.\n",
    "\n",
    "**Support** (also called range) is the interval over which the values of the data set spread. Since GAPs are positive and must be no larger that 4, we know that the range must be a subset of the interval [0, 4]. Presumably, students will GPAs lower than 2 will not apply to a graduate program, so the support of our GPA data will likely be a subset of the interval [2, 4].\n",
    "\n",
    "**Mode** is the most common value in the data set.  \n",
    "\n",
    "**Median** is the value that is larger than half of all values and smaller than half of all values in the data set. The median is an example of a *percentile*.  Two other common percentiles are the *first quartile* and the *third quartile*.\n",
    "\n",
    "**Sample Mean** (also called sample average) is the sum of all values divided by the number of observations.  The sample mean has the smallest distance to the set of all values in the sample.\n",
    "\n",
    "**Standard deviation** is a measure of the spread around the sample mean for the values in the data set.\n",
    "\n",
    "**Skewness** is a measure of the asymmetry of the values in the data set. If you divide the support of the data at the sample mean, and if one of the interval is longer than the other, than the data is skewed.\n",
    "\n",
    "These quantities can all be easily obtained used methods already coded in `Scipy` and `Numpy`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:52:20.497616Z",
     "start_time": "2019-07-02T14:52:20.484189Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mode of the sample is 2.00\n",
      "The first quartile of the sample is -2.00\n",
      "The median of the sample is 1.00\n",
      "The third quartile of the sample is 2.00\n",
      "The skewness of the sample is -0.415\n"
     ]
    }
   ],
   "source": [
    "#import scipy.stats as stat\n",
    "#import numpy as np\n",
    "\n",
    "mode = float(stat.mode(othello_sentiment)[0])\n",
    "print(\"The mode of the sample is {0:4.2f}\".format(mode))\n",
    "\n",
    "first_quartile = stat.scoreatpercentile(othello_sentiment, 25)\n",
    "print(\"The first quartile of the sample is {0:4.2f}\".format(first_quartile))\n",
    "\n",
    "median = stat.scoreatpercentile(othello_sentiment, 50)\n",
    "print(\"The median of the sample is {0:4.2f}\".format(median))\n",
    "\n",
    "third_quartile = stat.scoreatpercentile(othello_sentiment, 75)\n",
    "print(\"The third quartile of the sample is {0:4.2f}\".format(third_quartile))\n",
    "\n",
    "skew = stat.skew(othello_sentiment)\n",
    "print(\"The skewness of the sample is {0:5.3f}\".format(skew))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:52:20.924499Z",
     "start_time": "2019-07-02T14:52:20.916580Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DescribeResult(nobs=470, minmax=(-4, 4), mean=0.4085106382978723, variance=5.120609717370594, skewness=-0.41543139442792626, kurtosis=-1.3060163694576852)\n",
      "\n",
      "DescribeResult(nobs=537, minmax=(-4, 4), mean=0.32774674115456237, variance=4.907307040217906, skewness=-0.29978164467913276, kurtosis=-1.4307035152922654)\n"
     ]
    }
   ],
   "source": [
    "results = stat.describe(othello_sentiment)\n",
    "print(results)\n",
    "\n",
    "print()\n",
    "\n",
    "results = stat.describe(iago_sentiment)\n",
    "print(results)\n",
    "\n",
    "# Let's prettify these results so that they are easier to read by humans\n",
    "# You code here!\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot the distribution\n",
    "\n",
    "While descriptive statistics are very useful, their calculation involves the loss of a lot of information on the data.  Obtaining an histogram of the data values gives a much more accurate picture of the statistical properties of the data *as long as the histogram is calculated properly*. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-01T23:00:15.016678Z",
     "start_time": "2019-07-01T23:00:14.548101Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lgaalves/miniconda3/lib/python3.6/site-packages/matplotlib/axes/_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x648 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure( figsize = (8, 9) )\n",
    "my_font_size = 15\n",
    "\n",
    "sub1 = fig.add_subplot(2,1,1)\n",
    "sub2 = fig.add_subplot(2,1,2)\n",
    "\n",
    "half_frame(sub1, \"\", \"Probability mass\", font_size = my_font_size)\n",
    "half_frame(sub2, \"Sentiment score\", \"Cumulative probability\", font_size = my_font_size)\n",
    "\n",
    "# Calculate and plot histogram\n",
    "#\n",
    "sub1.hist([othello_sentiment, iago_sentiment], 9, normed = 1, color = ['r', 'darkblue'], rwidth = 0.75, alpha = 0.5, \n",
    "          histtype = \"bar\", label = ['Othello', 'Iago'], cumulative = False)\n",
    "\n",
    "sub2.hist([othello_sentiment, iago_sentiment], 9, normed = 1, color = ['r', 'darkblue'], rwidth = 0.75, alpha = 0.5, \n",
    "          histtype = \"bar\", label = ['Othello', 'Iago'], cumulative = True)\n",
    "\n",
    "# Format legend\n",
    "sub1.legend(loc = \"best\", frameon = False, markerscale = 1.8, fontsize = my_font_size)\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What question we may want to answer with our data is whether Othello and Iago express similar levels of sentiment in their speeches.  Looking at the probability mass function, one could be tempted to say maybe... The cumulative probability plot, however, suggest that probably not.  How do we decide?  \n",
    "\n",
    "\n",
    "## Inferential statistics\n",
    "\n",
    "The goal of descriptive statistics is to gain sufficient insight into our data to enable us to develop hypotheses about what process generated the data or whether different samples were drawn from the same distribution. **These hypotheses are called null models**.\n",
    "\n",
    "The process can be a physical model, which may or may not predict the value of the parameters necessary to describe the statistical properties of the data. The process can also be a statistical model.\n",
    "\n",
    "Using inferential statistics, one determines whether a given null hypothesis is consistent with the data. Let's test the null hypothesis using a few standard methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:53:31.794710Z",
     "start_time": "2019-07-02T14:53:31.789238Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t test: \t t =  0.5706  \t p = 0.5684\n"
     ]
    }
   ],
   "source": [
    "# Using t-test\n",
    "t, p = stat.ttest_ind(othello_sentiment, iago_sentiment, equal_var = False)\n",
    "print('t test: \\t t =  {0:6.4f}  \\t p = {1:6.4f}'.format(t, p) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:54:17.984199Z",
     "start_time": "2019-07-02T14:54:17.976154Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KS 2 sample test: \t  D =  0.0272  \t p = 0.9916\n"
     ]
    }
   ],
   "source": [
    "# Using Kolmogorov-Smirnov test\n",
    "D, p = stat.ks_2samp(othello_sentiment, iago_sentiment)\n",
    "print('KS 2 sample test: \\t  D =  {0:6.4f}  \\t p = {1:6.4f}'.format(D, p) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-02T14:55:03.827661Z",
     "start_time": "2019-07-02T14:55:03.813370Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 17  43  88  37   0  61 136  83   5]\n",
      "[ 12  49 117  48   0  67 149  92   3]\n",
      "chi^2 goodness of fit test: \t chi^2 =  4.1771  \t p = 0.8993722457702136\n"
     ]
    }
   ],
   "source": [
    "# Using chi_square test\n",
    "# http://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/chi2samp.htm\n",
    "\n",
    "hist_othello, edges = np.histogram(othello_sentiment, bins = 9, density = False)\n",
    "hist_iago, edges = np.histogram(iago_sentiment, bins = 9, density = False)\n",
    "print(hist_othello)\n",
    "print(hist_iago)\n",
    "\n",
    "# Calculate chisquared statistic\n",
    "chisquared = 0\n",
    "factor1 = math.sqrt(N_iago/N_othello)\n",
    "factor2 = math.sqrt(N_othello/N_iago)\n",
    "for i in range(len(hist_othello)):\n",
    "    if hist_othello[i] != 0:\n",
    "        chisquared += (factor1 * hist_othello[i] - factor2 * hist_iago[i])**2 / ((hist_othello[i] + hist_iago[i]))\n",
    "\n",
    "p = stat.distributions.chi2.sf(chisquared, len(hist_othello))\n",
    "print('chi^2 goodness of fit test: \\t chi^2 =  {0:6.4f}  \\t p = {1}'.format(chisquared, p) )\n",
    "chisquared_data = chisquared"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok!  So all three tests suggest that the differences are not significant. Should we trust these results, though?\n",
    "\n",
    "If we think about it, the `t-test` and the `KS` are **clearly** not good tests for these data. The data is bimodal (there goes the `t-test`) and discrete/categorical  (there goes the `KS test`).  What about the $\\chi^2$ test? How do we know whether its assumptions are fullfilled?\n",
    "\n",
    "The $\\chi^2$ test requires three conditions to be satisfied: 1) the data is obtained by simple random sampling, 2) the data is categorical, and 3) the expected number of observations for every category is at least 5.\n",
    "\n",
    "Conditions 2 and 3 are clearly fullfilled. Condition 1 is a little trickier. We are not really doing any sampling. We are getting all the data that there is to get except for the possibility that the play was altered during or after its creation and that the removed parts were select in a biased manner... \n",
    "\n",
    "What do we do in case it is really difficult to figure out whether the conditions of a test are fullfilled?\n",
    "\n",
    "# Bootstrap methods\n",
    "\n",
    "That is where bootstrapping approaches can help!!\n",
    "\n",
    "We want to test the null hypothesis that the two samples are drawn from the same distribution.  Well, if both samples are drawn from the same distribution, we can put all the data together into a single sample.  And then we can use that combined sample, to generate pretty much an unlimited number of synthetic samples with the same sizes as our data.\n",
    "\n",
    "Let's do that!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-03T11:08:06.800809Z",
     "start_time": "2019-07-03T11:08:06.794800Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1007\n"
     ]
    }
   ],
   "source": [
    "# Generate combined data set\n",
    "\n",
    "joint_sample = copy(othello_sentiment)\n",
    "joint_sample.extend(iago_sentiment)\n",
    "print(len(joint_sample))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-03T11:10:15.871626Z",
     "start_time": "2019-07-03T11:10:15.860213Z"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-03T10:59:17.624459Z",
     "start_time": "2019-07-03T10:59:17.615551Z"
    }
   },
   "outputs": [],
   "source": [
    "# Generate synthetic data sets\n",
    "othello_synthetic = sample(joint_sample, N_othello)\n",
    "iago_synthetic = sample(joint_sample, N_iago)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-03T11:10:36.949495Z",
     "start_time": "2019-07-03T11:10:36.923115Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ttest_ind - data: \t t =  0.5706\n",
      "ttest_ind - synthetic: \t t =  -0.8153\n"
     ]
    }
   ],
   "source": [
    "# Use t-test to check differences in average\n",
    "t, p = stat.ttest_ind(othello_sentiment, iago_sentiment, equal_var = False)\n",
    "print('ttest_ind - data: \\t t =  {0:6.4f}'.format(t) )\n",
    "\n",
    "t, p = stat.ttest_ind(othello_synthetic, iago_synthetic, equal_var = False)\n",
    "print('ttest_ind - synthetic: \\t t =  {0:6.4f}'.format(t) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-03T11:25:30.237075Z",
     "start_time": "2019-07-03T11:25:30.003003Z"
    }
   },
   "outputs": [],
   "source": [
    "# Write code to perform these analyses repeatedly\n",
    "N_replicates = 100\n",
    "t_statistic = []\n",
    "D_statistic = []\n",
    "chisquared_statistic = []\n",
    "\n",
    "for i in range(0,N_replicates):\n",
    "    # Generate random samples\n",
    "    othello_synthetic = sample(joint_sample, N_othello)\n",
    "    iago_synthetic = sample(joint_sample, N_iago)\n",
    "    \n",
    "    # t-student test\n",
    "    t, p = stat.ttest_ind(othello_synthetic, iago_synthetic, equal_var = False)\n",
    "    t_statistic.append(t)\n",
    "    \n",
    "    # Kolmogorov-Smirnov test\n",
    "    D, p = stat.ks_2samp(othello_synthetic, iago_synthetic)\n",
    "    D_statistic.append(D)\n",
    "    \n",
    "    # Calculate chisquared statistic\n",
    "    hist_othello, edges = np.histogram(othello_synthetic, bins = 9, density = False)\n",
    "    hist_iago, edges = np.histogram(iago_synthetic, bins = 9, density = False)\n",
    "\n",
    "    chisquared = 0\n",
    "    factor1 = math.sqrt(N_iago/N_othello)\n",
    "    factor2 = math.sqrt(N_othello/N_iago)\n",
    "    for i in range(len(hist_othello)):\n",
    "        if hist_othello[i] != 0:\n",
    "            chisquared += (factor1 * hist_othello[i] - factor2 * hist_iago[i])**2 / ((hist_othello[i] + hist_iago[i]))\n",
    "    chisquared_statistic.append(chisquared)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once, we have our results, we can plot them"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-07-03T11:25:33.422275Z",
     "start_time": "2019-07-03T11:25:32.277733Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x1080 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure( figsize = (8, 15) )\n",
    "my_font_size = 15\n",
    "\n",
    "sub1 = fig.add_subplot(3,1,1)\n",
    "sub2 = fig.add_subplot(3,1,2)\n",
    "sub3 = fig.add_subplot(3,1,3)\n",
    "\n",
    "half_frame(sub1, \"t statistic\", \"Probability density\", font_size = my_font_size)\n",
    "half_frame(sub2, \"D statistic\", \"Probability density\", font_size = my_font_size)\n",
    "half_frame(sub3, \"chi^2 statistic\", \"Probability density\", font_size = my_font_size)\n",
    "\n",
    "# Calculate and plot histogram\n",
    "#\n",
    "sub1.hist(t_statistic, 30, normed = 1, color = 'darkblue', rwidth = 0.75, alpha = 0.5, \n",
    "          histtype = \"bar\", cumulative = True)\n",
    "\n",
    "t, p = stat.ttest_ind(othello_sentiment, iago_sentiment, equal_var = False)\n",
    "sub1.vlines(t, ymin = 0, ymax = 0.99, lw = 3, color = \"red\", label = \"data\")\n",
    "\n",
    "sub2.hist(D_statistic, 30, normed = 1, color = 'darkblue', rwidth = 0.75, alpha = 0.5, \n",
    "          histtype = \"bar\", cumulative = True)\n",
    "\n",
    "D, p = stat.ks_2samp(othello_sentiment, iago_sentiment)\n",
    "sub2.vlines(D, ymin = 0, ymax = 1, lw = 3, color = \"red\", label = \"data\")\n",
    "\n",
    "\n",
    "sub3.hist(chisquared_statistic, 200, normed = 1, color = 'darkblue', rwidth = 0.75, alpha = 0.5, \n",
    "          histtype = \"bar\", cumulative = True)\n",
    "\n",
    "sub3.vlines(chisquared_data, ymin = 0, ymax = 1, lw = 3, color = \"red\", label = \"data\")\n",
    "sub3.set_xlim([0,20])\n",
    "\n",
    "# Format legend\n",
    "#sub1.legend(loc = \"best\", frameon = False, markerscale = 1.8, fontsize = my_font_size)\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that the differences between the sentiment levels of Iago's and Othello's speech are not statistically significant.  It would be interesting to compare their speeches with those uttered by characters from other plays..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autoclose": false,
   "autocomplete": true,
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  },
  "toc": {
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}