Stats Notes.ipynb

Stats Notes.ipynb

{
  "cells": [
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import numpy as np",
      "execution_count": 1,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x = np.random.normal(size=100)",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x",
      "execution_count": 3,
      "outputs": [
        {
          "data": {
            "text/plain": "array([-0.64636638, -0.92067658,  0.57067851, -1.1747002 ,  0.23594245,\n       -0.28177888,  0.04127362,  0.21306925,  2.43165718, -0.77082127,\n       -0.26585945, -0.91241284, -1.64701622, -0.29050484, -1.19140542,\n        0.93300485, -0.83420451, -0.44973678, -1.62028674,  0.1859824 ,\n       -0.02606027,  0.37424804, -0.89680962, -0.93087077,  0.21047056,\n       -1.5271699 ,  0.47173263, -0.18185374, -0.04670623, -0.73831363,\n       -1.81337616, -0.35674319,  0.6133092 , -0.84690014, -1.36129476,\n       -0.98962838,  2.40653107,  0.43185476,  0.3021308 , -0.74235255,\n       -1.43230168, -0.51865809,  0.80809757, -0.72772726,  1.07503972,\n       -0.45384097, -0.11555224,  1.32794185,  0.15541308,  1.13780852,\n       -0.99337909,  0.27535055, -0.36047248, -0.40394354, -2.93613591,\n        0.0189053 , -0.83085372,  0.81773645, -1.87821709, -0.4777017 ,\n        0.21790097,  1.76329118,  0.58818108,  0.48495416,  1.05951994,\n       -0.92796168,  0.34885423,  0.10624113,  0.25353643,  0.27338819,\n        0.11443639,  1.40443838, -0.71029859, -0.11426099,  0.75107925,\n       -0.37260427, -0.37812208, -0.43184358,  0.0605904 ,  0.45415155,\n       -0.89592852, -1.2448475 ,  0.77970559,  0.12195701, -0.86675726,\n       -2.1638069 , -1.68179993, -0.56914989, -0.28887394, -0.24832861,\n       -0.98082885,  0.18305305, -0.42348633,  0.72064558, -0.89348183,\n        0.24229539,  1.57774938, -0.15869812,  0.48114033,  1.96390124])"
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import matplotlib.pyplot as plt\n%matplotlib inline",
      "execution_count": 5,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plt.hist(x)",
      "execution_count": 6,
      "outputs": [
        {
          "data": {
            "text/plain": "(array([ 1.,  2.,  7., 17., 22., 24., 16.,  5.,  3.,  3.]),\n array([-2.93613591, -2.3993566 , -1.86257729, -1.32579798, -0.78901868,\n        -0.25223937,  0.28453994,  0.82131925,  1.35809856,  1.89487787,\n         2.43165718]),\n <a list of 10 Patch objects>)"
          },
          "execution_count": 6,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x109a8fcf8>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "epsilon = np.random.normal(size=100)\ny = x + epsilon",
      "execution_count": 7,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "epsilon",
      "execution_count": 8,
      "outputs": [
        {
          "data": {
            "text/plain": "array([ 0.24624381, -0.22676508,  0.07557954,  1.82262933, -0.35685628,\n       -0.19634871, -2.61946767,  0.64111686, -0.76184957, -0.02286739,\n       -0.38491345, -0.8982332 ,  1.45678462, -0.83435824, -0.40646613,\n       -1.86854325,  1.36901135,  1.31236656, -1.6695548 ,  1.14880158,\n        1.26202185, -0.6548988 ,  0.50580688,  0.81565079,  0.64046087,\n       -0.90062536, -1.72083968, -1.38307941,  0.2012319 , -2.23225782,\n        0.2814349 ,  1.48327444, -1.04629655, -0.55502442, -0.91338703,\n        0.26026754, -0.15432357,  0.28532621,  1.11749372, -0.73807323,\n       -1.88657601, -0.56871774, -0.11727565,  0.85418437,  0.15408943,\n       -0.03363813,  2.05884473,  1.91443572,  0.23777464, -0.24382372,\n       -1.45999834, -0.31893434,  0.63250797,  1.17906818, -0.10101611,\n        0.9602161 ,  1.81776968, -0.36402809,  1.05389745,  0.97646444,\n        0.54642906, -0.97252875, -1.32407229, -1.28524841, -0.11596952,\n       -0.08377689, -0.77453557, -0.95488011,  1.6076321 ,  0.37045067,\n        0.32541772,  0.10935306,  0.29170895, -0.39898111, -0.4633239 ,\n       -1.50253645, -0.37415856,  0.43566928,  0.80312415, -0.55816526,\n        0.02086006,  0.64302759, -0.58702331, -0.35847869, -0.49427091,\n       -1.41322997,  0.40979316, -1.70699721, -0.88565058,  0.63753813,\n        0.08088755,  0.56829729,  0.94720841,  0.43149777,  1.78888479,\n       -1.07135318,  1.12715167,  1.02388739, -1.31348497,  0.24129767])"
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plt.plot(x, y, 'o')",
      "execution_count": 10,
      "outputs": [
        {
          "data": {
            "text/plain": "[<matplotlib.lines.Line2D at 0x10b85aef0>]"
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x10b791ba8>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "epsilon = np.random.normal(size=100)\ny = x +  1 * epsilon\nplt.plot(x, y, 'o')\nprint(np.corrcoef(x, y))",
      "execution_count": 29,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": "[[1.         0.63587592]\n [0.63587592 1.        ]]\n"
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAD8CAYAAABjAo9vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAGZZJREFUeJzt3X2MXNV5x/Hfw7KQhShdKrZNWXBsqcjIhASrK0LlSg1OWtOQYIcobdLQKM0fbqREDQg5skNUEiURlqzmRU3+sQKKIixCUqjTBCoDNREtklHWscFQ45S8AUtaNg2bN2+CX57+MbtmPXvvzJ25576d+/1Iltjx+N4zY/zMmec85znm7gIAxOOMqgcAAAiLwA4AkSGwA0BkCOwAEBkCOwBEhsAOAJEhsANAZAjsABAZAjsARObMUBcysxFJ05Jm3P2tvZ57/vnn+8qVK0PdGgBaYf/+/T9194l+zwsW2CV9WNJhSa/q98SVK1dqeno64K0BIH5m9uMszwuSijGzCyVdI+lLIa4HABheqBz75yR9RNLJQNcDAAwpd2A3s7dKesHd9/d53mYzmzaz6dnZ2by3BQCkCDFjXyfpWjP7kaSvSlpvZnd0P8ndd7r7lLtPTUz0zf0DAIaUO7C7+zZ3v9DdV0p6l6S97n597pEBAIYSsioGQAvtPjCjHXuO6Pm5eV0wPqYtG1Zr09rJqofVakEDu7t/W9K3Q14TQH3tPjCjbfcc0vyxE5Kkmbl5bbvnkCQR3CvEzlMAQ9ux58ipoL5o/tgJ7dhzpKIRQSIVAyCH5+fmB3q8DKSGmLEDyOGC8bGBHi/aYmpoZm5erpdTQ7sPzFQynqoQ2AEMbcuG1RobHTntsbHREW3ZsLqS8ZAa6iAVA2BoiymOuqQ+6pgaqgKBHUAum9ZO1iaHfcH4mGYSgnhVqaGqkIoBEI26pYaqwowdQDTqlhqqCoEdQFTqlBqqCqkYAIgMgR0AIkNgB4DIENgBIDIEdgCIDIEdACJDYAeAyFDHDgAFqqKNcO7AbmavkPSwpLMXrvfP7n5L3usCQNNVdcJUiFTMbyWtd/fXS7pc0tVmdmWA6wJAo1XVRjh3YPeOXy38OLrwy/NeFwCaLq1d8MzcfKGHfwRZPDWzETM7KOkFSQ+4+6MhrgsATdarXXCRJzsFCezufsLdL5d0oaQrzOy13c8xs81mNm1m07OzsyFuCwC1ltRGeFGRKZmg5Y7uPifpIUlXJ/zeTnefcvepiYmJkLcFgFratHZSt153WervF3WyU+7AbmYTZja+8N9jkv5M0lN5rwsAMdi0dlKTJR/6HWLG/geSHjKzxyV9R50c+7cCXBcAolD2yU6569jd/XFJawOMBQCiVPbJTuw8BYASlHmyE71iACAyBHYAiAyBHQAiQ2AHgMiweAqUrIo2rmgXAjtQoqrauKJdSMUAJaqqjSvahcAOlCitN0hRPUPQTgR2oERpvUGK6hmCdiKwAyVK6hli6uTa123fW+jhC1hu94EZrdu+V6u23hvV+8/iKVCipT1DZubmZXr5uDEWUssV80I2M3agZJvWTuqRres1OT627AxJFlLLE/NCNoEdqAgLqdWK+f0nsAMVYSG1WjG//wR2oCJlH76A08X8/rN4ilap03b+sg9fwOlifv/NvXv5ZsALmF0k6SuSfl+dBf6d7v75Xn9mamrKp6enc90XGFR3FYTUmaHdet1lUfxjRvzMbL+7T/V7XohUzHFJN7n7GklXSvqgma0JcF0gqJirIIClcgd2d/+Ju3934b9/KemwJKY/qJ2YqyCApYIunprZSnUOtn404fc2m9m0mU3Pzs6GvC2QScxVEMBSwQK7mb1S0t2SbnD3X3T/vrvvdPcpd5+amJgIdVsgsyZUQcS6xR3lClIVY2aj6gT1Xe5+T4hrAqHVvQoi5i3uKFfuwG5mJuk2SYfd/TP5hwQUZ9PaydoGyV6Lu3UdM+opRCpmnaS/kbTezA4u/HpLgOsCrcLiLkLJPWN39/9Up/MogBwuGB/TTEIQZ3EXg6KlAFATTVjcRTPQUgCoibov7qI5COxAjdR5cRfNQWAHKlKnhmSIC4EdqAA16ygSgR2oQBU163xDaA8CO1CBQWrWQwRkviG0C+WOQAWyNiRbDMgzc/NyvRyQB+0hQ8vidiGwAxXIWrMeKiCzq7VdCOxABTatndSt112myfExmaTJ8bHEk5xCBWRaFrcLOXagIllq1kO1GdiyYXXisYDsao0TM3agxkK1Gcj6DQFxyH2Y9TA4zBrIrogyRUofmynrYdakYoCaC91mgNLH+JGKAVqG0sf4MWNHa5B+6KD0MX7M2NEKoTb6xIDSx/gFCexmdruZvWBmT4S4HhBaXdMPuw/MaN32vVq19V6t2763lA8aDvSIX6hUzJclfUHSVwJdDwiqDumH7lTQVZdM6O79M6UvYnKgR/yCBHZ3f9jMVoa4FlCEqs8TTapE2bXvGXUXGxfd4XERB3rErbQcu5ltNrNpM5uenZ0t67aApOrTD0mpoLQdJCxiIq/SAru773T3KXefmpiYKOu2gKTqd14OEqxZxERelDuiNapMP6Slgkynz9xZxEQIlDsCGeStXklLBb3nyhX0b0FwQWbsZnanpDdKOt/MnpN0i7vfFuLaGBwbccIKsQWfShSUiSZgkekOQlJnZshMcHjrtu9NTKNMjo/pka3rSxlDrw9rPsjbgyZgLVXFIcmxS1v4nJmb16qt9xYeTHt9Y5BEQy8sQ2CPTB024sQmbeFT0mntCaRigmm/XbN8kKMbi6eRia0PSBVb7rslLXx2K7I9Qa8Paz7IkYTAHpmqN+KEVJfGXd018GmKCqa9Pqxj+yBHGKRiIhNT9cUnvvlkbdIMS2vg0xZTiwqm/c4rreIs07IXbFkgHgyBPUIx9AHZfWBGLx49lvh7VacZBjkYOkRAyvJhXXaQLXPBlhOfBke5I2opbVYshS0zHDbwZvlzsZaell3+WYdy07qg3BGN1mtWHirNkGcmmOVbUaylp2Uv2LJAPDgCO2oprcRwfGw0WFAsOvDGGpDytEBO6kn/0FOzPb/5VN1yuYmoikGpspYvplX3fPzaS4ONpejAG2vFyrCVV0lVTnfse6Zv1VNMlV5lIbCjNIOUL5bRZrfowJsnINWhfj/NsH83Sd+QuiXtB6i65XITsXiK0tRtEayMxc1hFmdjXXRdtfXe1MNFljJJP9x+TdHDaSQWT1E7dcs5l1HzP0zpaayLrr1aM3Q/D/kQ2FGaOi6C1bHmv24fgKEk1f93I3ceBjl2lKbsRbA656l7iXXRNSlXfj0HjRSCGTtKU2a7gzrvVuyXdx9kZ2vT1PEbUowI7ChVWf+w8+api+pNkuUDJ6Z+P6hGqKPxrpb0eUkjkr7k7ttDXBcYVp48dZGz/awfOGkfgDTDQha5c+xmNiLpi5L+QtIaSe82szV5rwvkkSdP3e9gi0XD5PBDfOBU3cYY9Rdi8fQKSU+7+w/c/SVJX5W0McB1gaHlWajNEnyHDbJlfOAAIQL7pKRnl/z83MJjpzGzzWY2bWbTs7OzAW4LpMuzWzFL8B02yBb9gQNIJS6euvtOSTulzs7Tsu6L5sqbTx52oTZLVcqwQTbPwmgd9wGgnkLM2GckXbTk5wsXHgOGVmU+OctsP09KZdPaSW3ZsFoXjI/p+bl57dhzJNProhkWsgoxY/+OpIvNbJU6Af1dkv46wHXRYmVuq0/7ZtDrPnlOUbrqkgndvX9m4KobyiCRVe7A7u7HzexDkvaoU+54u7s/mXtkaLWy8sn9Shv7Bf1BT1GamZvXrn3PLGuGlfVDiw0+yCJIjt3d75N0X4hroTp1qpEuK5/cbxG0V9DvDrK7D8zo8k/cr7n5zlmt550zKnctu37aAlNTFkHr9P8JktErBpLqVyNdVj651zeDtKB/09ceW/a+7D4woy1ff+xUUJekF48eO+3nfpqwCFq3/0+QjMAOSfWrkS7rcIVei6BpQf+E+7JgtmPPER07OXyx1+iINWIRtG7/nyAZvWIgqdycdtav8WXkk3stgu7YcyS1f3h3TnyQ92l0xHTihOvk0gczfCaEToEMcz1q6ZuBwA5J5eS069hxsd8iaK/+4UuDWa9DJM47Z1TnnHXmqev/+rfHl6Vojp103fS1x3TjXQcTg2zo927YRWNq6ZuBwA5J5bSKrevJQGnfDBYfu+lrj+lEwhGSS4PZlg2rteXrjy1Lx4yOmG5526WnXX/V1nsTx7F4j6SgHfq9G3bROOaWwjEhxw5J5eS0m/g1ftPaSf3jX76+70LuprWT2vHO12t8bPTUY+eeNaJzzzpTN9518LQmYVlmt91567T3KMtRc0mGWTRe/BDhYOn6Y8aOU4rOaTf1a3zWmvWl799iqmMx5dJv1pskS6rHFu416N9br7+Lfh/A1NLXHzN2ZBLimLmkEkZJOvrS8dLK5YZ9HZvWTuqRrev1w+3X6JGt6/sGtkFmvSNmidfoTvUkPcsX7jWoXuWksR7N1ybM2NFXnoW77kW4d/zRpL712E+W1XuXsYja63VIYbfqDzLr7R6XlJzqueGugwPdq5dBF43JozcLgR19DbtwlxRI794/o7PPXP5Fcf7YCd1w10Ht2HOksJ2Maa/jE998Ur85djJotc4gaaesqZ7JwKmsfovG7C5tLgI7+hp20TMtkPbKLRdZApk23hePLt8dmrda56pLJnTHvmcSH0+SJW9dZkUKefRmI8eOvobNuQ5b7VLUTsZBZ7Z5qnUeeir5MJm0x7OgIgVZMWNHX8POFNPSEeedM3pa6iNJESWQaa/j7DPPSOzpkmexsKjSTmbSyIIZO/oadqaYVnlxy9suPXW9NEVUYKS9jo9fe2nwhmO9vuWEqDACemHGjkyGmSn2W4Rb3LpeRt64uzrns391+bLXE3KxMO3bwVWXTNSurQLiY56wVbpoU1NTPj09Xfp9UU9F9/dO+/AoOj+d9LrSGotNjo/pka3rg9+PD4u4mNl+d5/q+zwCO+omdIBat31vYcF0UKu23pvYyNEk/XD7NUNft6oPL5Qra2DPlWM3s3ea2ZNmdtLM+t4M6KeIgxzq1KOmqF2d9EnHUnkXT5+QdJ2khwOMBQ1S1AJgvwA1zH3rtEW+qJOh6vThherlWjx198OSZCm9LhCnIvuq9wpQw963V7lm2XnponZ1NrXBGopRWlWMmW2WtFmSVqxYUdZtUYAi+6r3ClD97psWpNOCqdT7sOqiFFGLTp90LNU3sJvZg5JenfBbN7v7N7LeyN13StopdRZPM48QtVPk1/5eAerGHk2w+s3mk4Lpuu17a3nwxzDo74Kl+gZ2d39zGQNBcxT5tb9XgEorFcwym08SW16aXalYxM5TDKyoBcBFab3Pe9231wlDaYusdVpULRM7X+OXt9zx7Wb2nKQ/lnSvme0JMyzUWYhmVMMEl1737RWM00omi/6AqqMiyklRP2xQQumK2EyTdM1uSRuS2rZbs06btTC4rBuU6BWD0oWsqlkamMfPGU3t1Cgl587blpeObV0ByQjsKF3SjLHX42m6Z+kvHj2msdERjY+NBm/Dm2UsTZj5U+/eDiyeonRphzenPZ4mbeZvplJz503KW7dxXaGNCOwo3YmUdZ20x9OkpQ/mjh47bZF1fGxUrxg9QzfedbCQKpAm9WnhFKZ2IBWD0p13zmjiOaO9Dt5I0iutsJg7L7L9waKseeu6pGvatq7QRszYUardB2b0q98cX/b46IgNnA7IklYoYzadpR6+SekaNB+BHaXaseeIjp1cnnI596wzhzqhqV9aoYwqkLp8wACLSMWgcB/bfUh3Pvpszxz6z1NKFPvpl1YoowokS58WygxRJgI7CvWx3Yd0x75n+j7vd8ZGtW773uD557K6HtbhAwZYRComAnXu/XHno8/2fc7oGaZfv3S8kPxzXapAKDNEmZixN1wZVR959Eq/mDoz1qMvHV9WJROyfW4dqkBoq4syEdgbrshDL0IYMUsM7iNm+v6tb5HUOeA5SWz55zp8wKAdSMU0XN0X5d79hov6Pt7W9rlAUQjsDVf3oPipTZfp+itXnGoXMGKm669coU9tuuzUc8g/A2GRimm4Jpx1+alNl50WyJda3I05f+zEqbTNZIH557rs/gSK1JjAzj/IZFkX5er4/nUv/J5wP/WhVFRQr/NCMxBKroM2zGyHpLdJeknS9yX9rbvP9ftzgx60UcTBDG1S1/ev7EMfOGQCTZf1oI28OfYHJL3W3V8n6XuStuW8XqI2bscOWZte1/ev7IXfui80A6HkCuzufr+7L3Z02ifpwvxDWq5t/yBDN4yq6/tX9sJv3ReagVBCVsW8X9K/BbzeKW37Bxl6hl3X96/sahiqb9AWfQO7mT1oZk8k/Nq45Dk3SzouaVeP62w2s2kzm56dnR1okG37Bxl6hl3X96/o7f7d6SxJtWgvABQt1+KpJJnZ+yT9naQ3ufvRLH9m0MVTqZ5VHUUpYpGvTe+fVN8FYyCPrIuneatirpb0GUl/6u6Zp+HDBPY2ISjlRwUMYpQ1sOetY/+CpLMlPWCdnYX73P0DOa/ZejSM6sjzLaOuC8ZAGXIFdnf/w1ADwena3jAq72Yi+p+jzegVg1rKWxlU1wVjoAyNaSmAdsmbSiGddbq2LZ63HYEdtRQildL2dNYieuS0D6mYhqjT8XdljIVUSjh1bSmB4jBjb4A6zbjKGguplHCoEGofAnsD1On4uzLHQiolDCqE2odUTAPUacZVp7EgG9Ja7UNgb4A6NfGq01iQTdE9eVA/pGIaoE7H39VpLHhZv3JG0lrtQmBvgDotJNZpLOio0+I66iF3d8dh0ASsvdgoEx4Nz9qjrCZgaLCyg2zazHL6xz/TQ0/NEuyHxII2uhHYW6qKr+9ppZK79j2jxe+NocfRhm8IlDOiG1UxLVXFbsS0GWR3MjDUOEKfHVtXlDOiG4G9par4+j7IDDLEONqylZ5yRnQjFdNSVXx9TyqVNC2fsYcaR5tyz5QzYilm7C1Vxdf3pJnle65cUdg4Bt1MVadGa0AeuWbsZvZJSRslnZT0gqT3ufvzIQaGYlVVj540s5x6ze8WMo5BNlNRC46Y5D3M+lXu/ouF//57SWuynHlKHTvKkrUqhlpwNEEpdeyLQX3BuUpOl6IkbSjtG1TW3HOb8vGIX+7FUzP7tKT3Svq5pKtyjwhDIZWQD7XgiEnfxVMze9DMnkj4tVGS3P1md79I0i5JH+pxnc1mNm1m07Ozs+FeASS1p7SvKNSCIyZ9Z+zu/uaM19ol6T5Jt6RcZ6eknVInx551gMiGVEI+NDdDTPJWxVzs7v+98ONGSU/lHxKGQSohP2rBEYu8dezbF9Iyj0v6c0kfDjAmDIFUAoBFeati3hFqIMiHVAKARbQUiAipBAASLQUAIDoEdgCIDIEdACJDYAeAyBDYASAyBHYAiAyBHQAiQ2AHgMgQ2AEgMgR2AIgMgR0AIkNgB4DIENgBIDIEdgCIDG17C7L7wAy90QFUgsBegN0HZrTtnkOnDpeemZvXtnsOSRLBHUDhgqRizOwmM3MzOz/E9Zpux54jp4L6ovljJ7Rjz5GKRgSgTXIHdjO7SJ3zTp/JP5w4PJ9wqHSvxwEgpBAz9s9K+ogkD3CtKFwwPjbQ4wAQUq7AbmYbJc24+2OBxhOFLRtWa2x05LTHxkZHtGXD6opGBKBN+i6emtmDkl6d8Fs3S/qoOmmYvsxss6TNkrRixYoBhtg8iwukVMUAqIK5D5dBMbPLJP27pKMLD10o6XlJV7j7//T6s1NTUz49PT3UfQGgrcxsv7tP9Xve0OWO7n5I0u8tueGPJE25+0+HvSYAID92ngJAZIJtUHL3laGuBQAYHjN2AIgMgR0AIjN0VUyum5rNSvrxAH/kfEkxL8rG/vokXmMMYn99Uv1f42vcfaLfkyoJ7IMys+ksJT5NFfvrk3iNMYj99UnxvEZSMQAQGQI7AESmKYF9Z9UDKFjsr0/iNcYg9tcnRfIaG5FjBwBk15QZOwAgo0YEdjP7pJk9bmYHzex+M7ug6jGFZmY7zOyphdf5L2Y2XvWYQjOzd5rZk2Z20swaX3mwyMyuNrMjZva0mW2tejyhmdntZvaCmT1R9ViKYmYXmdlDZvZfC/+PfrjqMeXRiMAuaYe7v87dL5f0LUn/UPWACvCApNe6++skfU/StorHU4QnJF0n6eGqBxKKmY1I+qKkv5C0RtK7zWxNtaMK7suSrq56EAU7Lukmd18j6UpJH2zy32MjAru7/2LJj+cqwtOa3P1+dz++8OM+ddogR8XdD7t7bAe/XiHpaXf/gbu/JOmrkjZWPKag3P1hST+rehxFcvefuPt3F/77l5IOS2rsAQrBmoAVzcw+Lem9kn4u6aqKh1O090u6q+pBIJNJSc8u+fk5SW+oaCwIwMxWSlor6dFqRzK82gT2Xic1ufs33P1mSTeb2TZJH5J0S6kDDKDfa1x4zs3qfC3cVebYQsnyGoG6MrNXSrpb0g1dmYJGqU1gd/c3Z3zqLkn3qYGBvd9rNLP3SXqrpDd5Q+tQB/h7jMWMpIuW/HzhwmNoGDMbVSeo73L3e6oeTx6NyLGb2cVLftwo6amqxlIUM7ta0kckXevuR/s9H7XxHUkXm9kqMztL0rsk/WvFY8KAzMwk3SbpsLt/purx5NWIDUpmdrek1ZJOqtMV8gPuHtWsyMyelnS2pP9beGifu3+gwiEFZ2Zvl/RPkiYkzUk66O4bqh1Vfmb2FkmfkzQi6XZ3/3TFQwrKzO6U9EZ1Oh/+r6Rb3P22SgcVmJn9iaT/kHRInTgjSR919/uqG9XwGhHYAQDZNSIVAwDIjsAOAJEhsANAZAjsABAZAjsARIbADgCRIbADQGQI7AAQmf8Hh4r6hBbN5TsAAAAASUVORK5CYII=\n",
            "text/plain": "<matplotlib.figure.Figure at 0x10c81e860>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import pandas as pd",
      "execution_count": 30,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "z = x + y + epsilon",
      "execution_count": 31,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df = pd.DataFrame({'x': x, 'y': y, 'z': z})",
      "execution_count": 32,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.corr()",
      "execution_count": 34,
      "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>x</th>\n      <th>y</th>\n      <th>z</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>x</th>\n      <td>1.000000</td>\n      <td>0.635876</td>\n      <td>0.635876</td>\n    </tr>\n    <tr>\n      <th>y</th>\n      <td>0.635876</td>\n      <td>1.000000</td>\n      <td>1.000000</td>\n    </tr>\n    <tr>\n      <th>z</th>\n      <td>0.635876</td>\n      <td>1.000000</td>\n      <td>1.000000</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
            "text/plain": "          x         y         z\nx  1.000000  0.635876  0.635876\ny  0.635876  1.000000  1.000000\nz  0.635876  1.000000  1.000000"
          },
          "execution_count": 34,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "pd.plotting.scatter_matrix(df, figsize=(10, 10));",
      "execution_count": 37,
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x1103796d8>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df = pd.read_csv(\"https://raw.githubusercontent.com/rpruim/OpenIntro/master/data/hsb2.csv\")",
      "execution_count": 39,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "pd.plotting.scatter_matrix(df[['read', 'write', 'math', 'science', 'socst']], \n                           figsize=(12, 12));",
      "execution_count": 43,
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x110606320>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df[['read', 'write', 'math', 'science', 'socst']].corr()",
      "execution_count": 44,
      "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>read</th>\n      <th>write</th>\n      <th>math</th>\n      <th>science</th>\n      <th>socst</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>read</th>\n      <td>1.000000</td>\n      <td>0.596776</td>\n      <td>0.662280</td>\n      <td>0.630158</td>\n      <td>0.621484</td>\n    </tr>\n    <tr>\n      <th>write</th>\n      <td>0.596776</td>\n      <td>1.000000</td>\n      <td>0.617449</td>\n      <td>0.570442</td>\n      <td>0.604793</td>\n    </tr>\n    <tr>\n      <th>math</th>\n      <td>0.662280</td>\n      <td>0.617449</td>\n      <td>1.000000</td>\n      <td>0.630733</td>\n      <td>0.544480</td>\n    </tr>\n    <tr>\n      <th>science</th>\n      <td>0.630158</td>\n      <td>0.570442</td>\n      <td>0.630733</td>\n      <td>1.000000</td>\n      <td>0.465106</td>\n    </tr>\n    <tr>\n      <th>socst</th>\n      <td>0.621484</td>\n      <td>0.604793</td>\n      <td>0.544480</td>\n      <td>0.465106</td>\n      <td>1.000000</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
            "text/plain": "             read     write      math   science     socst\nread     1.000000  0.596776  0.662280  0.630158  0.621484\nwrite    0.596776  1.000000  0.617449  0.570442  0.604793\nmath     0.662280  0.617449  1.000000  0.630733  0.544480\nscience  0.630158  0.570442  0.630733  1.000000  0.465106\nsocst    0.621484  0.604793  0.544480  0.465106  1.000000"
          },
          "execution_count": 44,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from statsmodels.formula.api import ols",
      "execution_count": 45,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "ols('science ~ math', data=df).fit().summary()",
      "execution_count": 47,
      "outputs": [
        {
          "data": {
            "text/html": "<table class=\"simpletable\">\n<caption>OLS Regression Results</caption>\n<tr>\n  <th>Dep. Variable:</th>         <td>science</td>     <th>  R-squared:         </th> <td>   0.398</td>\n</tr>\n<tr>\n  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.395</td>\n</tr>\n<tr>\n  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   130.8</td>\n</tr>\n<tr>\n  <th>Date:</th>             <td>Thu, 22 Mar 2018</td> <th>  Prob (F-statistic):</th> <td>1.39e-23</td>\n</tr>\n<tr>\n  <th>Time:</th>                 <td>19:12:47</td>     <th>  Log-Likelihood:    </th> <td> -691.09</td>\n</tr>\n<tr>\n  <th>No. Observations:</th>      <td>   200</td>      <th>  AIC:               </th> <td>   1386.</td>\n</tr>\n<tr>\n  <th>Df Residuals:</th>          <td>   198</td>      <th>  BIC:               </th> <td>   1393.</td>\n</tr>\n<tr>\n  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n</tr>\n<tr>\n  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n</tr>\n<tr>\n  <th>Intercept</th> <td>   16.7579</td> <td>    3.116</td> <td>    5.378</td> <td> 0.000</td> <td>   10.613</td> <td>   22.903</td>\n</tr>\n<tr>\n  <th>math</th>      <td>    0.6666</td> <td>    0.058</td> <td>   11.437</td> <td> 0.000</td> <td>    0.552</td> <td>    0.782</td>\n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n  <th>Omnibus:</th>       <td> 0.997</td> <th>  Durbin-Watson:     </th> <td>   1.878</td>\n</tr>\n<tr>\n  <th>Prob(Omnibus):</th> <td> 0.608</td> <th>  Jarque-Bera (JB):  </th> <td>   0.665</td>\n</tr>\n<tr>\n  <th>Skew:</th>          <td> 0.082</td> <th>  Prob(JB):          </th> <td>   0.717</td>\n</tr>\n<tr>\n  <th>Kurtosis:</th>      <td> 3.230</td> <th>  Cond. No.          </th> <td>    306.</td>\n</tr>\n</table>",
            "text/plain": "<class 'statsmodels.iolib.summary.Summary'>\n\"\"\"\n                            OLS Regression Results                            \n==============================================================================\nDep. Variable:                science   R-squared:                       0.398\nModel:                            OLS   Adj. R-squared:                  0.395\nMethod:                 Least Squares   F-statistic:                     130.8\nDate:                Thu, 22 Mar 2018   Prob (F-statistic):           1.39e-23\nTime:                        19:12:47   Log-Likelihood:                -691.09\nNo. Observations:                 200   AIC:                             1386.\nDf Residuals:                     198   BIC:                             1393.\nDf Model:                           1                                         \nCovariance Type:            nonrobust                                         \n==============================================================================\n                 coef    std err          t      P>|t|      [0.025      0.975]\n------------------------------------------------------------------------------\nIntercept     16.7579      3.116      5.378      0.000      10.613      22.903\nmath           0.6666      0.058     11.437      0.000       0.552       0.782\n==============================================================================\nOmnibus:                        0.997   Durbin-Watson:                   1.878\nProb(Omnibus):                  0.608   Jarque-Bera (JB):                0.665\nSkew:                           0.082   Prob(JB):                        0.717\nKurtosis:                       3.230   Cond. No.                         306.\n==============================================================================\n\nWarnings:\n[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n\"\"\""
          },
          "execution_count": 47,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "$$science = 0.6666 \\times math + 16.75$$"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df['math'].min()",
      "execution_count": 49,
      "outputs": [
        {
          "data": {
            "text/plain": "33"
          },
          "execution_count": 49,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "0.630733 ** 2",
      "execution_count": 50,
      "outputs": [
        {
          "data": {
            "text/plain": "0.39782411728899997"
          },
          "execution_count": 50,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "alcohol = np.random.uniform(size=100)",
      "execution_count": 51,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "alcohol",
      "execution_count": 52,
      "outputs": [
        {
          "data": {
            "text/plain": "array([0.55114607, 0.99198662, 0.32668944, 0.26266552, 0.42318107,\n       0.95713339, 0.19187847, 0.38706297, 0.39929302, 0.61769309,\n       0.45933433, 0.25228983, 0.38205216, 0.47549067, 0.17173173,\n       0.99276646, 0.7957686 , 0.62265087, 0.48558226, 0.83436406,\n       0.33681353, 0.42250122, 0.16209433, 0.59580854, 0.02768948,\n       0.92730631, 0.82932744, 0.44222019, 0.0898678 , 0.80500555,\n       0.69217801, 0.09598253, 0.7683574 , 0.29535633, 0.777758  ,\n       0.12497139, 0.25514578, 0.46531037, 0.28896541, 0.57257904,\n       0.64649636, 0.63855875, 0.95737833, 0.93224986, 0.51075404,\n       0.43775678, 0.99192219, 0.35004279, 0.6746585 , 0.41711445,\n       0.63394504, 0.75676451, 0.39166938, 0.51799876, 0.19117126,\n       0.21344247, 0.1290969 , 0.55046808, 0.28823591, 0.08758783,\n       0.6959017 , 0.58666909, 0.55808448, 0.96331724, 0.56290279,\n       0.96881882, 0.88826716, 0.43964954, 0.14010347, 0.07077454,\n       0.66822643, 0.8739614 , 0.66853682, 0.69508562, 0.02232407,\n       0.63058472, 0.65366518, 0.03217863, 0.91546377, 0.17232138,\n       0.94299881, 0.42037605, 0.43231934, 0.68093484, 0.26206191,\n       0.84896971, 0.59044137, 0.33778788, 0.41843326, 0.70407646,\n       0.96402916, 0.49067335, 0.05432362, 0.0669876 , 0.74032356,\n       0.24785911, 0.39878419, 0.45209055, 0.66235062, 0.57420416])"
          },
          "execution_count": 52,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "smoking = alcohol + np.random.uniform(size=100) * 0.5",
      "execution_count": 53,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "life_length = 100 - alcohol * 50 + np.random.uniform(low=-5, high=5, size=100)",
      "execution_count": 55,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df = pd.DataFrame({'alcohol': alcohol, 'smoking': smoking, \n                   'life_length': life_length})",
      "execution_count": 56,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.plot.scatter(x='smoking', y='life_length')",
      "execution_count": 59,
      "outputs": [
        {
          "data": {
            "text/plain": "<matplotlib.axes._subplots.AxesSubplot at 0x114a96c50>"
          },
          "execution_count": 59,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x1107ae7b8>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.plot.scatter(x='smoking', y='alcohol')",
      "execution_count": 60,
      "outputs": [
        {
          "data": {
            "text/plain": "<matplotlib.axes._subplots.AxesSubplot at 0x11496e6a0>"
          },
          "execution_count": 60,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x114c9e470>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.plot.scatter(y='life_length', x='alcohol')",
      "execution_count": 62,
      "outputs": [
        {
          "data": {
            "text/plain": "<matplotlib.axes._subplots.AxesSubplot at 0x111feada0>"
          },
          "execution_count": 62,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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\n",
            "text/plain": "<matplotlib.figure.Figure at 0x114a24be0>"
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df",
      "execution_count": 71,
      "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>alcohol</th>\n      <th>life_length</th>\n      <th>smoking</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.551146</td>\n      <td>72.097342</td>\n      <td>0.646734</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0.991987</td>\n      <td>48.449375</td>\n      <td>1.221210</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0.326689</td>\n      <td>87.232226</td>\n      <td>0.589601</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0.262666</td>\n      <td>87.847355</td>\n      <td>0.361461</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0.423181</td>\n      <td>81.000070</td>\n      <td>0.732635</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>0.957133</td>\n      <td>51.790827</td>\n      <td>1.009736</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>0.191878</td>\n      <td>94.620566</td>\n      <td>0.419861</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>0.387063</td>\n      <td>84.322231</td>\n      <td>0.496446</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>0.399293</td>\n      <td>75.400349</td>\n      <td>0.763963</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>0.617693</td>\n      <td>66.173432</td>\n      <td>0.901973</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>0.459334</td>\n      <td>76.641429</td>\n      <td>0.573136</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>0.252290</td>\n      <td>91.126505</td>\n      <td>0.606873</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0.382052</td>\n      <td>82.606919</td>\n      <td>0.395330</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0.475491</td>\n      <td>72.101267</td>\n      <td>0.733393</td>\n    </tr>\n    <tr>\n      <th>14</th>\n      <td>0.171732</td>\n      <td>87.088022</td>\n      <td>0.277942</td>\n    </tr>\n    <tr>\n      <th>15</th>\n      <td>0.992766</td>\n      <td>50.436298</td>\n      <td>1.048269</td>\n    </tr>\n    <tr>\n      <th>16</th>\n      <td>0.795769</td>\n      <td>55.343827</td>\n      <td>1.287561</td>\n    </tr>\n    <tr>\n      <th>17</th>\n      <td>0.622651</td>\n      <td>67.410327</td>\n      <td>0.624511</td>\n    </tr>\n    <tr>\n      <th>18</th>\n      <td>0.485582</td>\n      <td>71.067534</td>\n      <td>0.822940</td>\n    </tr>\n    <tr>\n      <th>19</th>\n      <td>0.834364</td>\n      <td>62.814027</td>\n      <td>1.107102</td>\n    </tr>\n    <tr>\n      <th>20</th>\n      <td>0.336814</td>\n      <td>85.526630</td>\n      <td>0.731327</td>\n    </tr>\n    <tr>\n      <th>21</th>\n      <td>0.422501</td>\n      <td>83.834179</td>\n      <td>0.901576</td>\n    </tr>\n    <tr>\n      <th>22</th>\n      <td>0.162094</td>\n      <td>88.104596</td>\n      <td>0.293803</td>\n    </tr>\n    <tr>\n      <th>23</th>\n      <td>0.595809</td>\n      <td>71.623192</td>\n      <td>0.738765</td>\n    </tr>\n    <tr>\n      <th>24</th>\n      <td>0.027689</td>\n      <td>101.286127</td>\n      <td>0.081993</td>\n    </tr>\n    <tr>\n      <th>25</th>\n      <td>0.927306</td>\n      <td>56.322282</td>\n      <td>1.253279</td>\n    </tr>\n    <tr>\n      <th>26</th>\n      <td>0.829327</td>\n      <td>53.570543</td>\n      <td>0.917489</td>\n    </tr>\n    <tr>\n      <th>27</th>\n      <td>0.442220</td>\n      <td>77.693484</td>\n      <td>0.936925</td>\n    </tr>\n    <tr>\n      <th>28</th>\n      <td>0.089868</td>\n      <td>96.051726</td>\n      <td>0.283131</td>\n    </tr>\n    <tr>\n      <th>29</th>\n      <td>0.805006</td>\n      <td>62.414501</td>\n      <td>1.183274</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>70</th>\n      <td>0.668226</td>\n      <td>65.773904</td>\n      <td>0.732876</td>\n    </tr>\n    <tr>\n      <th>71</th>\n      <td>0.873961</td>\n      <td>55.997877</td>\n      <td>1.085860</td>\n    </tr>\n    <tr>\n      <th>72</th>\n      <td>0.668537</td>\n      <td>67.156891</td>\n      <td>0.920700</td>\n    </tr>\n    <tr>\n      <th>73</th>\n      <td>0.695086</td>\n      <td>61.417821</td>\n      <td>1.106472</td>\n    </tr>\n    <tr>\n      <th>74</th>\n      <td>0.022324</td>\n      <td>102.624652</td>\n      <td>0.168808</td>\n    </tr>\n    <tr>\n      <th>75</th>\n      <td>0.630585</td>\n      <td>70.287438</td>\n      <td>0.655377</td>\n    </tr>\n    <tr>\n      <th>76</th>\n      <td>0.653665</td>\n      <td>62.345489</td>\n      <td>0.947507</td>\n    </tr>\n    <tr>\n      <th>77</th>\n      <td>0.032179</td>\n      <td>94.391888</td>\n      <td>0.220113</td>\n    </tr>\n    <tr>\n      <th>78</th>\n      <td>0.915464</td>\n      <td>54.753822</td>\n      <td>1.265414</td>\n    </tr>\n    <tr>\n      <th>79</th>\n      <td>0.172321</td>\n      <td>93.984679</td>\n      <td>0.274503</td>\n    </tr>\n    <tr>\n      <th>80</th>\n      <td>0.942999</td>\n      <td>51.843464</td>\n      <td>1.412112</td>\n    </tr>\n    <tr>\n      <th>81</th>\n      <td>0.420376</td>\n      <td>74.129856</td>\n      <td>0.710794</td>\n    </tr>\n    <tr>\n      <th>82</th>\n      <td>0.432319</td>\n      <td>80.635069</td>\n      <td>0.718175</td>\n    </tr>\n    <tr>\n      <th>83</th>\n      <td>0.680935</td>\n      <td>68.229179</td>\n      <td>1.167156</td>\n    </tr>\n    <tr>\n      <th>84</th>\n      <td>0.262062</td>\n      <td>87.578517</td>\n      <td>0.735764</td>\n    </tr>\n    <tr>\n      <th>85</th>\n      <td>0.848970</td>\n      <td>61.297730</td>\n      <td>0.905677</td>\n    </tr>\n    <tr>\n      <th>86</th>\n      <td>0.590441</td>\n      <td>67.263478</td>\n      <td>0.761981</td>\n    </tr>\n    <tr>\n      <th>87</th>\n      <td>0.337788</td>\n      <td>79.636977</td>\n      <td>0.455746</td>\n    </tr>\n    <tr>\n      <th>88</th>\n      <td>0.418433</td>\n      <td>75.719520</td>\n      <td>0.469911</td>\n    </tr>\n    <tr>\n      <th>89</th>\n      <td>0.704076</td>\n      <td>62.278214</td>\n      <td>0.753867</td>\n    </tr>\n    <tr>\n      <th>90</th>\n      <td>0.964029</td>\n      <td>55.218831</td>\n      <td>1.014514</td>\n    </tr>\n    <tr>\n      <th>91</th>\n      <td>0.490673</td>\n      <td>76.623406</td>\n      <td>0.494682</td>\n    </tr>\n    <tr>\n      <th>92</th>\n      <td>0.054324</td>\n      <td>95.352934</td>\n      <td>0.240466</td>\n    </tr>\n    <tr>\n      <th>93</th>\n      <td>0.066988</td>\n      <td>93.560140</td>\n      <td>0.422029</td>\n    </tr>\n    <tr>\n      <th>94</th>\n      <td>0.740324</td>\n      <td>64.606509</td>\n      <td>0.742938</td>\n    </tr>\n    <tr>\n      <th>95</th>\n      <td>0.247859</td>\n      <td>89.860328</td>\n      <td>0.603568</td>\n    </tr>\n    <tr>\n      <th>96</th>\n      <td>0.398784</td>\n      <td>83.608821</td>\n      <td>0.433673</td>\n    </tr>\n    <tr>\n      <th>97</th>\n      <td>0.452091</td>\n      <td>73.089579</td>\n      <td>0.840297</td>\n    </tr>\n    <tr>\n      <th>98</th>\n      <td>0.662351</td>\n      <td>67.027343</td>\n      <td>1.034622</td>\n    </tr>\n    <tr>\n      <th>99</th>\n      <td>0.574204</td>\n      <td>76.174755</td>\n      <td>0.660952</td>\n    </tr>\n  </tbody>\n</table>\n<p>100 rows × 3 columns</p>\n</div>",
            "text/plain": "     alcohol  life_length   smoking\n0   0.551146    72.097342  0.646734\n1   0.991987    48.449375  1.221210\n2   0.326689    87.232226  0.589601\n3   0.262666    87.847355  0.361461\n4   0.423181    81.000070  0.732635\n5   0.957133    51.790827  1.009736\n6   0.191878    94.620566  0.419861\n7   0.387063    84.322231  0.496446\n8   0.399293    75.400349  0.763963\n9   0.617693    66.173432  0.901973\n10  0.459334    76.641429  0.573136\n11  0.252290    91.126505  0.606873\n12  0.382052    82.606919  0.395330\n13  0.475491    72.101267  0.733393\n14  0.171732    87.088022  0.277942\n15  0.992766    50.436298  1.048269\n16  0.795769    55.343827  1.287561\n17  0.622651    67.410327  0.624511\n18  0.485582    71.067534  0.822940\n19  0.834364    62.814027  1.107102\n20  0.336814    85.526630  0.731327\n21  0.422501    83.834179  0.901576\n22  0.162094    88.104596  0.293803\n23  0.595809    71.623192  0.738765\n24  0.027689   101.286127  0.081993\n25  0.927306    56.322282  1.253279\n26  0.829327    53.570543  0.917489\n27  0.442220    77.693484  0.936925\n28  0.089868    96.051726  0.283131\n29  0.805006    62.414501  1.183274\n..       ...          ...       ...\n70  0.668226    65.773904  0.732876\n71  0.873961    55.997877  1.085860\n72  0.668537    67.156891  0.920700\n73  0.695086    61.417821  1.106472\n74  0.022324   102.624652  0.168808\n75  0.630585    70.287438  0.655377\n76  0.653665    62.345489  0.947507\n77  0.032179    94.391888  0.220113\n78  0.915464    54.753822  1.265414\n79  0.172321    93.984679  0.274503\n80  0.942999    51.843464  1.412112\n81  0.420376    74.129856  0.710794\n82  0.432319    80.635069  0.718175\n83  0.680935    68.229179  1.167156\n84  0.262062    87.578517  0.735764\n85  0.848970    61.297730  0.905677\n86  0.590441    67.263478  0.761981\n87  0.337788    79.636977  0.455746\n88  0.418433    75.719520  0.469911\n89  0.704076    62.278214  0.753867\n90  0.964029    55.218831  1.014514\n91  0.490673    76.623406  0.494682\n92  0.054324    95.352934  0.240466\n93  0.066988    93.560140  0.422029\n94  0.740324    64.606509  0.742938\n95  0.247859    89.860328  0.603568\n96  0.398784    83.608821  0.433673\n97  0.452091    73.089579  0.840297\n98  0.662351    67.027343  1.034622\n99  0.574204    76.174755  0.660952\n\n[100 rows x 3 columns]"
          },
          "execution_count": 71,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "$$lifelength = \\beta_0 + \\beta_{smoking} \\times smoking + \\beta_{alcohol} \\times alcohol + \\epsilon$$\n\nПусть $\\beta_{smoking} = -1$, $\\beta_{alcohol} = -2$."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "ols('life_length ~ smoking + alcohol', data=df).fit().summary()",
      "execution_count": 64,
      "outputs": [
        {
          "data": {
            "text/html": "<table class=\"simpletable\">\n<caption>OLS Regression Results</caption>\n<tr>\n  <th>Dep. Variable:</th>       <td>life_length</td>   <th>  R-squared:         </th> <td>   0.957</td>\n</tr>\n<tr>\n  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.956</td>\n</tr>\n<tr>\n  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   1072.</td>\n</tr>\n<tr>\n  <th>Date:</th>             <td>Thu, 22 Mar 2018</td> <th>  Prob (F-statistic):</th> <td>7.36e-67</td>\n</tr>\n<tr>\n  <th>Time:</th>                 <td>19:50:33</td>     <th>  Log-Likelihood:    </th> <td> -248.78</td>\n</tr>\n<tr>\n  <th>No. Observations:</th>      <td>   100</td>      <th>  AIC:               </th> <td>   503.6</td>\n</tr>\n<tr>\n  <th>Df Residuals:</th>          <td>    97</td>      <th>  BIC:               </th> <td>   511.4</td>\n</tr>\n<tr>\n  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n</tr>\n<tr>\n  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n</tr>\n<tr>\n  <th>Intercept</th> <td>  100.3386</td> <td>    0.827</td> <td>  121.361</td> <td> 0.000</td> <td>   98.698</td> <td>  101.979</td>\n</tr>\n<tr>\n  <th>smoking</th>   <td>   -1.4997</td> <td>    2.037</td> <td>   -0.736</td> <td> 0.463</td> <td>   -5.542</td> <td>    2.542</td>\n</tr>\n<tr>\n  <th>alcohol</th>   <td>  -48.5997</td> <td>    2.256</td> <td>  -21.543</td> <td> 0.000</td> <td>  -53.077</td> <td>  -44.122</td>\n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n  <th>Omnibus:</th>       <td>21.482</td> <th>  Durbin-Watson:     </th> <td>   1.889</td>\n</tr>\n<tr>\n  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>   5.221</td>\n</tr>\n<tr>\n  <th>Skew:</th>          <td>-0.109</td> <th>  Prob(JB):          </th> <td>  0.0735</td>\n</tr>\n<tr>\n  <th>Kurtosis:</th>      <td> 1.902</td> <th>  Cond. No.          </th> <td>    13.9</td>\n</tr>\n</table>",
            "text/plain": "<class 'statsmodels.iolib.summary.Summary'>\n\"\"\"\n                            OLS Regression Results                            \n==============================================================================\nDep. Variable:            life_length   R-squared:                       0.957\nModel:                            OLS   Adj. R-squared:                  0.956\nMethod:                 Least Squares   F-statistic:                     1072.\nDate:                Thu, 22 Mar 2018   Prob (F-statistic):           7.36e-67\nTime:                        19:50:33   Log-Likelihood:                -248.78\nNo. Observations:                 100   AIC:                             503.6\nDf Residuals:                      97   BIC:                             511.4\nDf Model:                           2                                         \nCovariance Type:            nonrobust                                         \n==============================================================================\n                 coef    std err          t      P>|t|      [0.025      0.975]\n------------------------------------------------------------------------------\nIntercept    100.3386      0.827    121.361      0.000      98.698     101.979\nsmoking       -1.4997      2.037     -0.736      0.463      -5.542       2.542\nalcohol      -48.5997      2.256    -21.543      0.000     -53.077     -44.122\n==============================================================================\nOmnibus:                       21.482   Durbin-Watson:                   1.889\nProb(Omnibus):                  0.000   Jarque-Bera (JB):                5.221\nSkew:                          -0.109   Prob(JB):                       0.0735\nKurtosis:                       1.902   Cond. No.                         13.9\n==============================================================================\n\nWarnings:\n[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n\"\"\""
          },
          "execution_count": 64,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "ols('life_length ~ alcohol', data=df).fit().summary()",
      "execution_count": 72,
      "outputs": [
        {
          "data": {
            "text/html": "<table class=\"simpletable\">\n<caption>OLS Regression Results</caption>\n<tr>\n  <th>Dep. Variable:</th>       <td>life_length</td>   <th>  R-squared:         </th> <td>   0.956</td>\n</tr>\n<tr>\n  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.956</td>\n</tr>\n<tr>\n  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   2153.</td>\n</tr>\n<tr>\n  <th>Date:</th>             <td>Thu, 22 Mar 2018</td> <th>  Prob (F-statistic):</th> <td>1.65e-68</td>\n</tr>\n<tr>\n  <th>Time:</th>                 <td>20:16:08</td>     <th>  Log-Likelihood:    </th> <td> -249.06</td>\n</tr>\n<tr>\n  <th>No. Observations:</th>      <td>   100</td>      <th>  AIC:               </th> <td>   502.1</td>\n</tr>\n<tr>\n  <th>Df Residuals:</th>          <td>    98</td>      <th>  BIC:               </th> <td>   507.3</td>\n</tr>\n<tr>\n  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n</tr>\n<tr>\n  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n</tr>\n<tr>\n  <th>Intercept</th> <td>   99.9437</td> <td>    0.628</td> <td>  159.195</td> <td> 0.000</td> <td>   98.698</td> <td>  101.190</td>\n</tr>\n<tr>\n  <th>alcohol</th>   <td>  -50.0576</td> <td>    1.079</td> <td>  -46.401</td> <td> 0.000</td> <td>  -52.199</td> <td>  -47.917</td>\n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n  <th>Omnibus:</th>       <td>24.155</td> <th>  Durbin-Watson:     </th> <td>   1.920</td>\n</tr>\n<tr>\n  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>   5.558</td>\n</tr>\n<tr>\n  <th>Skew:</th>          <td>-0.133</td> <th>  Prob(JB):          </th> <td>  0.0621</td>\n</tr>\n<tr>\n  <th>Kurtosis:</th>      <td> 1.876</td> <th>  Cond. No.          </th> <td>    4.68</td>\n</tr>\n</table>",
            "text/plain": "<class 'statsmodels.iolib.summary.Summary'>\n\"\"\"\n                            OLS Regression Results                            \n==============================================================================\nDep. Variable:            life_length   R-squared:                       0.956\nModel:                            OLS   Adj. R-squared:                  0.956\nMethod:                 Least Squares   F-statistic:                     2153.\nDate:                Thu, 22 Mar 2018   Prob (F-statistic):           1.65e-68\nTime:                        20:16:08   Log-Likelihood:                -249.06\nNo. Observations:                 100   AIC:                             502.1\nDf Residuals:                      98   BIC:                             507.3\nDf Model:                           1                                         \nCovariance Type:            nonrobust                                         \n==============================================================================\n                 coef    std err          t      P>|t|      [0.025      0.975]\n------------------------------------------------------------------------------\nIntercept     99.9437      0.628    159.195      0.000      98.698     101.190\nalcohol      -50.0576      1.079    -46.401      0.000     -52.199     -47.917\n==============================================================================\nOmnibus:                       24.155   Durbin-Watson:                   1.920\nProb(Omnibus):                  0.000   Jarque-Bera (JB):                5.558\nSkew:                          -0.133   Prob(JB):                       0.0621\nKurtosis:                       1.876   Cond. No.                         4.68\n==============================================================================\n\nWarnings:\n[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n\"\"\""
          },
          "execution_count": 72,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x = np.random.normal(size=10)\ny = np.random.normal(size=10)",
      "execution_count": 65,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x",
      "execution_count": 66,
      "outputs": [
        {
          "data": {
            "text/plain": "array([-0.12025737, -0.02251593, -0.93928455, -0.70500637,  1.05688788,\n       -0.65636919, -2.29201739,  1.06326959,  1.49651605,  0.29355656])"
          },
          "execution_count": 66,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "y",
      "execution_count": 67,
      "outputs": [
        {
          "data": {
            "text/plain": "array([-0.3652435 ,  1.30285537, -0.13253878,  0.94934658, -0.00630658,\n       -0.03706127, -1.04689335, -1.65468841,  1.18064708,  0.70962236])"
          },
          "execution_count": 67,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "np.corrcoef(x, y)",
      "execution_count": 68,
      "outputs": [
        {
          "data": {
            "text/plain": "array([[1.        , 0.22067453],\n       [0.22067453, 1.        ]])"
          },
          "execution_count": 68,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "xydf = pd.DataFrame({'x': x, 'y': y})",
      "execution_count": 69,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "ols('y ~ x', data=xydf).fit().summary()",
      "execution_count": 70,
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": "/usr/local/lib/python3.6/site-packages/scipy/stats/stats.py:1334: UserWarning: kurtosistest only valid for n>=20 ... continuing anyway, n=10\n  \"anyway, n=%i\" % int(n))\n"
        },
        {
          "data": {
            "text/html": "<table class=\"simpletable\">\n<caption>OLS Regression Results</caption>\n<tr>\n  <th>Dep. Variable:</th>            <td>y</td>        <th>  R-squared:         </th> <td>   0.049</td>\n</tr>\n<tr>\n  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>  -0.070</td>\n</tr>\n<tr>\n  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>  0.4095</td>\n</tr>\n<tr>\n  <th>Date:</th>             <td>Thu, 22 Mar 2018</td> <th>  Prob (F-statistic):</th>  <td> 0.540</td> \n</tr>\n<tr>\n  <th>Time:</th>                 <td>19:56:09</td>     <th>  Log-Likelihood:    </th> <td> -13.069</td>\n</tr>\n<tr>\n  <th>No. Observations:</th>      <td>    10</td>      <th>  AIC:               </th> <td>   30.14</td>\n</tr>\n<tr>\n  <th>Df Residuals:</th>          <td>     8</td>      <th>  BIC:               </th> <td>   30.74</td>\n</tr>\n<tr>\n  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n</tr>\n<tr>\n  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n</tr>\n<tr>\n  <th>Intercept</th> <td>    0.1055</td> <td>    0.317</td> <td>    0.333</td> <td> 0.748</td> <td>   -0.626</td> <td>    0.837</td>\n</tr>\n<tr>\n  <th>x</th>         <td>    0.1879</td> <td>    0.294</td> <td>    0.640</td> <td> 0.540</td> <td>   -0.489</td> <td>    0.865</td>\n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n  <th>Omnibus:</th>       <td> 1.925</td> <th>  Durbin-Watson:     </th> <td>   2.103</td>\n</tr>\n<tr>\n  <th>Prob(Omnibus):</th> <td> 0.382</td> <th>  Jarque-Bera (JB):  </th> <td>   0.723</td>\n</tr>\n<tr>\n  <th>Skew:</th>          <td>-0.657</td> <th>  Prob(JB):          </th> <td>   0.697</td>\n</tr>\n<tr>\n  <th>Kurtosis:</th>      <td> 2.906</td> <th>  Cond. No.          </th> <td>    1.11</td>\n</tr>\n</table>",
            "text/plain": "<class 'statsmodels.iolib.summary.Summary'>\n\"\"\"\n                            OLS Regression Results                            \n==============================================================================\nDep. Variable:                      y   R-squared:                       0.049\nModel:                            OLS   Adj. R-squared:                 -0.070\nMethod:                 Least Squares   F-statistic:                    0.4095\nDate:                Thu, 22 Mar 2018   Prob (F-statistic):              0.540\nTime:                        19:56:09   Log-Likelihood:                -13.069\nNo. Observations:                  10   AIC:                             30.14\nDf Residuals:                       8   BIC:                             30.74\nDf Model:                           1                                         \nCovariance Type:            nonrobust                                         \n==============================================================================\n                 coef    std err          t      P>|t|      [0.025      0.975]\n------------------------------------------------------------------------------\nIntercept      0.1055      0.317      0.333      0.748      -0.626       0.837\nx              0.1879      0.294      0.640      0.540      -0.489       0.865\n==============================================================================\nOmnibus:                        1.925   Durbin-Watson:                   2.103\nProb(Omnibus):                  0.382   Jarque-Bera (JB):                0.723\nSkew:                          -0.657   Prob(JB):                        0.697\nKurtosis:                       2.906   Cond. No.                         1.11\n==============================================================================\n\nWarnings:\n[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n\"\"\""
          },
          "execution_count": 70,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df_with_na = pd.DataFrame({'x': [1, 2, 3, np.nan, 2], \n                           'y': [2, 3, 1, 10,  np.nan]})",
      "execution_count": 77,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df_with_na",
      "execution_count": 78,
      "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>x</th>\n      <th>y</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1.0</td>\n      <td>2.0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2.0</td>\n      <td>3.0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>3.0</td>\n      <td>1.0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>NaN</td>\n      <td>10.0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2.0</td>\n      <td>NaN</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
            "text/plain": "     x     y\n0  1.0   2.0\n1  2.0   3.0\n2  3.0   1.0\n3  NaN  10.0\n4  2.0   NaN"
          },
          "execution_count": 78,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df_with_na.mean()",
      "execution_count": 81,
      "outputs": [
        {
          "data": {
            "text/plain": "x    2.0\ny    4.0\ndtype: float64"
          },
          "execution_count": 81,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df_with_na.fillna(df_with_na.mean())",
      "execution_count": 80,
      "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>x</th>\n      <th>y</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1.0</td>\n      <td>2.0</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2.0</td>\n      <td>3.0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>3.0</td>\n      <td>1.0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2.0</td>\n      <td>10.0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2.0</td>\n      <td>4.0</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
            "text/plain": "     x     y\n0  1.0   2.0\n1  2.0   3.0\n2  3.0   1.0\n3  2.0  10.0\n4  2.0   4.0"
          },
          "execution_count": 80,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from sklearn.decomposition import PCA",
      "execution_count": 82,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df",
      "execution_count": 83,
      "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>alcohol</th>\n      <th>life_length</th>\n      <th>smoking</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.551146</td>\n      <td>72.097342</td>\n      <td>0.646734</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0.991987</td>\n      <td>48.449375</td>\n      <td>1.221210</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0.326689</td>\n      <td>87.232226</td>\n      <td>0.589601</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0.262666</td>\n      <td>87.847355</td>\n      <td>0.361461</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>0.423181</td>\n      <td>81.000070</td>\n      <td>0.732635</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>0.957133</td>\n      <td>51.790827</td>\n      <td>1.009736</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>0.191878</td>\n      <td>94.620566</td>\n      <td>0.419861</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>0.387063</td>\n      <td>84.322231</td>\n      <td>0.496446</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>0.399293</td>\n      <td>75.400349</td>\n      <td>0.763963</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>0.617693</td>\n      <td>66.173432</td>\n      <td>0.901973</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>0.459334</td>\n      <td>76.641429</td>\n      <td>0.573136</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>0.252290</td>\n      <td>91.126505</td>\n      <td>0.606873</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0.382052</td>\n      <td>82.606919</td>\n      <td>0.395330</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0.475491</td>\n      <td>72.101267</td>\n      <td>0.733393</td>\n    </tr>\n    <tr>\n      <th>14</th>\n      <td>0.171732</td>\n      <td>87.088022</td>\n      <td>0.277942</td>\n    </tr>\n    <tr>\n      <th>15</th>\n      <td>0.992766</td>\n      <td>50.436298</td>\n      <td>1.048269</td>\n    </tr>\n    <tr>\n      <th>16</th>\n      <td>0.795769</td>\n      <td>55.343827</td>\n      <td>1.287561</td>\n    </tr>\n    <tr>\n      <th>17</th>\n      <td>0.622651</td>\n      <td>67.410327</td>\n      <td>0.624511</td>\n    </tr>\n    <tr>\n      <th>18</th>\n      <td>0.485582</td>\n      <td>71.067534</td>\n      <td>0.822940</td>\n    </tr>\n    <tr>\n      <th>19</th>\n      <td>0.834364</td>\n      <td>62.814027</td>\n      <td>1.107102</td>\n    </tr>\n    <tr>\n      <th>20</th>\n      <td>0.336814</td>\n      <td>85.526630</td>\n      <td>0.731327</td>\n    </tr>\n    <tr>\n      <th>21</th>\n      <td>0.422501</td>\n      <td>83.834179</td>\n      <td>0.901576</td>\n    </tr>\n    <tr>\n      <th>22</th>\n      <td>0.162094</td>\n      <td>88.104596</td>\n      <td>0.293803</td>\n    </tr>\n    <tr>\n      <th>23</th>\n      <td>0.595809</td>\n      <td>71.623192</td>\n      <td>0.738765</td>\n    </tr>\n    <tr>\n      <th>24</th>\n      <td>0.027689</td>\n      <td>101.286127</td>\n      <td>0.081993</td>\n    </tr>\n    <tr>\n      <th>25</th>\n      <td>0.927306</td>\n      <td>56.322282</td>\n      <td>1.253279</td>\n    </tr>\n    <tr>\n      <th>26</th>\n      <td>0.829327</td>\n      <td>53.570543</td>\n      <td>0.917489</td>\n    </tr>\n    <tr>\n      <th>27</th>\n      <td>0.442220</td>\n      <td>77.693484</td>\n      <td>0.936925</td>\n    </tr>\n    <tr>\n      <th>28</th>\n      <td>0.089868</td>\n      <td>96.051726</td>\n      <td>0.283131</td>\n    </tr>\n    <tr>\n      <th>29</th>\n      <td>0.805006</td>\n      <td>62.414501</td>\n      <td>1.183274</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>70</th>\n      <td>0.668226</td>\n      <td>65.773904</td>\n      <td>0.732876</td>\n    </tr>\n    <tr>\n      <th>71</th>\n      <td>0.873961</td>\n      <td>55.997877</td>\n      <td>1.085860</td>\n    </tr>\n    <tr>\n      <th>72</th>\n      <td>0.668537</td>\n      <td>67.156891</td>\n      <td>0.920700</td>\n    </tr>\n    <tr>\n      <th>73</th>\n      <td>0.695086</td>\n      <td>61.417821</td>\n      <td>1.106472</td>\n    </tr>\n    <tr>\n      <th>74</th>\n      <td>0.022324</td>\n      <td>102.624652</td>\n      <td>0.168808</td>\n    </tr>\n    <tr>\n      <th>75</th>\n      <td>0.630585</td>\n      <td>70.287438</td>\n      <td>0.655377</td>\n    </tr>\n    <tr>\n      <th>76</th>\n      <td>0.653665</td>\n      <td>62.345489</td>\n      <td>0.947507</td>\n    </tr>\n    <tr>\n      <th>77</th>\n      <td>0.032179</td>\n      <td>94.391888</td>\n      <td>0.220113</td>\n    </tr>\n    <tr>\n      <th>78</th>\n      <td>0.915464</td>\n      <td>54.753822</td>\n      <td>1.265414</td>\n    </tr>\n    <tr>\n      <th>79</th>\n      <td>0.172321</td>\n      <td>93.984679</td>\n      <td>0.274503</td>\n    </tr>\n    <tr>\n      <th>80</th>\n      <td>0.942999</td>\n      <td>51.843464</td>\n      <td>1.412112</td>\n    </tr>\n    <tr>\n      <th>81</th>\n      <td>0.420376</td>\n      <td>74.129856</td>\n      <td>0.710794</td>\n    </tr>\n    <tr>\n      <th>82</th>\n      <td>0.432319</td>\n      <td>80.635069</td>\n      <td>0.718175</td>\n    </tr>\n    <tr>\n      <th>83</th>\n      <td>0.680935</td>\n      <td>68.229179</td>\n      <td>1.167156</td>\n    </tr>\n    <tr>\n      <th>84</th>\n      <td>0.262062</td>\n      <td>87.578517</td>\n      <td>0.735764</td>\n    </tr>\n    <tr>\n      <th>85</th>\n      <td>0.848970</td>\n      <td>61.297730</td>\n      <td>0.905677</td>\n    </tr>\n    <tr>\n      <th>86</th>\n      <td>0.590441</td>\n      <td>67.263478</td>\n      <td>0.761981</td>\n    </tr>\n    <tr>\n      <th>87</th>\n      <td>0.337788</td>\n      <td>79.636977</td>\n      <td>0.455746</td>\n    </tr>\n    <tr>\n      <th>88</th>\n      <td>0.418433</td>\n      <td>75.719520</td>\n      <td>0.469911</td>\n    </tr>\n    <tr>\n      <th>89</th>\n      <td>0.704076</td>\n      <td>62.278214</td>\n      <td>0.753867</td>\n    </tr>\n    <tr>\n      <th>90</th>\n      <td>0.964029</td>\n      <td>55.218831</td>\n      <td>1.014514</td>\n    </tr>\n    <tr>\n      <th>91</th>\n      <td>0.490673</td>\n      <td>76.623406</td>\n      <td>0.494682</td>\n    </tr>\n    <tr>\n      <th>92</th>\n      <td>0.054324</td>\n      <td>95.352934</td>\n      <td>0.240466</td>\n    </tr>\n    <tr>\n      <th>93</th>\n      <td>0.066988</td>\n      <td>93.560140</td>\n      <td>0.422029</td>\n    </tr>\n    <tr>\n      <th>94</th>\n      <td>0.740324</td>\n      <td>64.606509</td>\n      <td>0.742938</td>\n    </tr>\n    <tr>\n      <th>95</th>\n      <td>0.247859</td>\n      <td>89.860328</td>\n      <td>0.603568</td>\n    </tr>\n    <tr>\n      <th>96</th>\n      <td>0.398784</td>\n      <td>83.608821</td>\n      <td>0.433673</td>\n    </tr>\n    <tr>\n      <th>97</th>\n      <td>0.452091</td>\n      <td>73.089579</td>\n      <td>0.840297</td>\n    </tr>\n    <tr>\n      <th>98</th>\n      <td>0.662351</td>\n      <td>67.027343</td>\n      <td>1.034622</td>\n    </tr>\n    <tr>\n      <th>99</th>\n      <td>0.574204</td>\n      <td>76.174755</td>\n      <td>0.660952</td>\n    </tr>\n  </tbody>\n</table>\n<p>100 rows × 3 columns</p>\n</div>",
            "text/plain": "     alcohol  life_length   smoking\n0   0.551146    72.097342  0.646734\n1   0.991987    48.449375  1.221210\n2   0.326689    87.232226  0.589601\n3   0.262666    87.847355  0.361461\n4   0.423181    81.000070  0.732635\n5   0.957133    51.790827  1.009736\n6   0.191878    94.620566  0.419861\n7   0.387063    84.322231  0.496446\n8   0.399293    75.400349  0.763963\n9   0.617693    66.173432  0.901973\n10  0.459334    76.641429  0.573136\n11  0.252290    91.126505  0.606873\n12  0.382052    82.606919  0.395330\n13  0.475491    72.101267  0.733393\n14  0.171732    87.088022  0.277942\n15  0.992766    50.436298  1.048269\n16  0.795769    55.343827  1.287561\n17  0.622651    67.410327  0.624511\n18  0.485582    71.067534  0.822940\n19  0.834364    62.814027  1.107102\n20  0.336814    85.526630  0.731327\n21  0.422501    83.834179  0.901576\n22  0.162094    88.104596  0.293803\n23  0.595809    71.623192  0.738765\n24  0.027689   101.286127  0.081993\n25  0.927306    56.322282  1.253279\n26  0.829327    53.570543  0.917489\n27  0.442220    77.693484  0.936925\n28  0.089868    96.051726  0.283131\n29  0.805006    62.414501  1.183274\n..       ...          ...       ...\n70  0.668226    65.773904  0.732876\n71  0.873961    55.997877  1.085860\n72  0.668537    67.156891  0.920700\n73  0.695086    61.417821  1.106472\n74  0.022324   102.624652  0.168808\n75  0.630585    70.287438  0.655377\n76  0.653665    62.345489  0.947507\n77  0.032179    94.391888  0.220113\n78  0.915464    54.753822  1.265414\n79  0.172321    93.984679  0.274503\n80  0.942999    51.843464  1.412112\n81  0.420376    74.129856  0.710794\n82  0.432319    80.635069  0.718175\n83  0.680935    68.229179  1.167156\n84  0.262062    87.578517  0.735764\n85  0.848970    61.297730  0.905677\n86  0.590441    67.263478  0.761981\n87  0.337788    79.636977  0.455746\n88  0.418433    75.719520  0.469911\n89  0.704076    62.278214  0.753867\n90  0.964029    55.218831  1.014514\n91  0.490673    76.623406  0.494682\n92  0.054324    95.352934  0.240466\n93  0.066988    93.560140  0.422029\n94  0.740324    64.606509  0.742938\n95  0.247859    89.860328  0.603568\n96  0.398784    83.608821  0.433673\n97  0.452091    73.089579  0.840297\n98  0.662351    67.027343  1.034622\n99  0.574204    76.174755  0.660952\n\n[100 rows x 3 columns]"
          },
          "execution_count": 83,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df = pd.read_csv(\"https://raw.githubusercontent.com/rpruim/OpenIntro/master/data/hsb2.csv\")",
      "execution_count": 84,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from sklearn.decomposition import PCA",
      "execution_count": 86,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "pca = PCA()",
      "execution_count": 87,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.columns",
      "execution_count": 88,
      "outputs": [
        {
          "data": {
            "text/plain": "Index(['id', 'gender', 'race', 'ses', 'schtyp', 'prog', 'read', 'write',\n       'math', 'science', 'socst'],\n      dtype='object')"
          },
          "execution_count": 88,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "pca.fit_transform(df[['read', 'write', 'math', 'science', 'socst']])",
      "execution_count": 89,
      "outputs": [
        {
          "data": {
            "text/plain": "array([[ 2.87766970e+00, -8.83700396e+00,  4.77087869e+00,\n        -3.49960266e+00,  9.06719090e+00],\n       [-1.92590733e+01,  7.48208325e-01,  5.83164985e+00,\n        -4.56139635e+00,  7.71231999e+00],\n       [ 1.92115378e+01,  2.12700165e+01,  7.69332428e+00,\n        -1.62857249e+00, -8.56036221e+00],\n       [-1.30503935e+00, -1.96669070e+00,  1.42036755e+01,\n        -3.13537466e+00,  4.68616091e+00],\n       [-3.53829354e+00, -5.09742615e+00, -1.78526589e+00,\n        -2.79389519e+00, -8.52172376e+00],\n       [-3.83378330e+00, -9.07888700e-01, -3.23646475e+00,\n        -1.38931998e+01, -7.08648113e+00],\n       [-1.57600906e+00, -9.17208073e+00, -4.51658653e+00,\n        -9.44727617e+00,  5.99788500e+00],\n       [ 2.81925834e+01,  2.25312159e+00, -7.43486875e+00,\n         3.30519553e+00, -2.19685411e+00],\n       [-9.55290328e+00,  5.46247890e+00,  3.03561682e+00,\n         8.83348553e-01,  6.66214410e+00],\n       [-1.51742308e+00, -1.84557314e-02,  1.03708304e+00,\n         3.09179743e+00,  4.53634314e+00],\n       [-4.75360070e+00, -5.15926748e+00,  1.11898218e+01,\n        -3.12101388e+00, -9.20561299e-02],\n       [-1.56153359e+01, -1.02999834e+00, -5.44346883e+00,\n        -8.17923673e+00,  5.62807223e+00],\n       [-3.36037573e+01, -3.29778381e+00,  8.25531369e+00,\n         5.40480589e+00, -3.54204150e+00],\n       [-5.35317903e+00,  7.10582452e+00, -7.94157902e+00,\n         4.37821544e+00,  4.02608487e+00],\n       [ 1.01021295e+01, -1.65468571e+01, -7.14962894e+00,\n         9.87907020e+00,  4.48724898e-01],\n       [ 9.74116966e+00, -6.64872689e+00, -2.28135480e+00,\n        -9.00347973e+00, -1.70671980e+00],\n       [ 2.65694374e+00, -4.54843831e+00, -2.04914820e+00,\n        -2.53561339e+00, -2.93770154e+00],\n       [-1.15539036e+01,  2.71264775e+00, -3.68875517e-01,\n         8.54340319e-01, -1.93105185e+00],\n       [-2.21760538e+01, -2.22372601e+00,  5.95280592e-01,\n         6.81346252e+00,  5.09471544e+00],\n       [ 5.15046240e+00,  7.95507195e+00,  1.13880679e+01,\n         2.71338767e+00, -5.64580194e+00],\n       [ 3.17183469e+00,  1.37278768e+01,  9.19699245e+00,\n        -1.31865388e+01,  1.77454632e+01],\n       [-3.40925904e+01,  7.01764698e+00, -1.02304292e+00,\n        -7.48441440e-01, -9.89834356e+00],\n       [ 6.66829697e+00, -1.89531825e+00,  9.51331959e+00,\n        -1.00971199e+01, -2.99208626e+00],\n       [-1.32701934e+01,  5.55174532e-01,  6.16595331e+00,\n        -4.30851668e+00, -2.13249330e+00],\n       [ 2.28663590e+01, -4.97835752e+00, -2.80916586e+00,\n         6.65720674e-01, -4.19011483e+00],\n       [ 3.39250080e+01,  7.04003062e+00, -1.09066085e+00,\n         4.26032913e+00, -5.36916501e+00],\n       [-2.73581027e+01, -1.71687944e+00, -8.45995324e-01,\n         2.47945390e+00, -1.07201534e+00],\n       [-7.20068491e-01,  8.85829158e+00, -7.52718457e+00,\n        -5.95797567e-01, -2.64854005e+00],\n       [ 2.03103679e+01, -2.99250614e+00,  5.90466183e+00,\n         4.21603796e+00, -5.59999366e+00],\n       [ 9.38955945e+00,  1.03312134e+01,  5.52242312e+00,\n        -1.76890425e+00,  1.34614190e+00],\n       [ 1.50406391e+01,  3.27869329e-01,  1.02992266e+01,\n         1.30286017e-01, -1.70542285e+01],\n       [-2.52791734e+01,  5.21329282e+00,  1.55187151e+01,\n         2.45168142e+00,  1.72909993e+00],\n       [-3.14350477e+01, -3.48234421e+00, -1.59916514e+00,\n        -3.97455428e+00, -5.21295130e-02],\n       [ 9.51567257e+00,  2.94695834e+01,  8.04389931e-02,\n        -8.91791884e+00, -9.37688425e+00],\n       [-9.37814338e+00, -2.32837647e+00, -3.75867597e+00,\n        -9.72661453e+00,  1.13599713e+00],\n       [-2.15399647e+01, -4.06166197e+00, -3.41862376e+00,\n        -3.57303090e+00,  1.45786128e+00],\n       [-3.01380003e+01,  4.57522612e+00,  9.03778142e+00,\n         3.58223784e+00, -1.05470563e+01],\n       [-2.90891995e+01,  1.97070181e+00,  1.38486687e+00,\n        -1.92023980e-01, -4.12021291e+00],\n       [ 2.99412392e+01,  2.39670958e+00,  1.22558194e+01,\n         6.60959970e+00, -6.87119607e-01],\n       [ 2.71485503e+01, -8.82022996e+00,  9.47007527e+00,\n        -5.00694867e+00, -6.82524548e+00],\n       [ 1.48087418e+01, -7.53111230e+00,  1.81703891e+00,\n        -1.85031514e+00,  3.54461472e+00],\n       [-9.91992828e+00,  6.70690131e+00, -3.88685983e+00,\n         3.29877803e+00, -2.08781150e+00],\n       [-2.94520291e+00,  3.88970157e+00, -1.52390142e+00,\n         8.97031036e-01, -2.34203679e+00],\n       [ 2.01937308e+01,  5.41944181e+00,  2.10830738e+00,\n        -5.21249514e+00, -1.02659404e+00],\n       [-2.05742189e+01, -5.00265049e+00,  2.36364029e-01,\n        -9.10869144e+00,  1.05781355e+01],\n       [-1.27112688e+01,  1.72437787e+01, -4.54137385e+00,\n        -1.96426579e+00, -2.32150811e+00],\n       [ 1.60393278e+01, -7.02281509e-02,  8.74956564e+00,\n        -6.49680221e+00,  3.95825332e+00],\n       [-2.03072413e+01,  1.23592314e+01,  1.74787102e+00,\n         7.46376469e+00, -2.32778660e+00],\n       [-1.96386057e+00,  7.23646617e+00, -9.45713305e+00,\n         6.48432907e+00, -5.28857210e+00],\n       [-2.77892407e+00,  3.10541846e-02,  1.39175186e+00,\n         8.96778172e+00, -2.33686485e-01],\n       [-1.13000687e+01,  1.75686373e+00, -5.99287501e+00,\n         1.45084535e+00, -2.94931322e+00],\n       [ 2.84345441e+01,  2.84673602e+00,  6.81168763e+00,\n        -8.02962367e+00, -2.57971421e+00],\n       [ 7.95731355e+00,  5.97834297e+00,  5.27521395e-01,\n         4.40314292e+00, -1.18431712e+01],\n       [-1.64202173e+01, -1.29304717e+01,  4.23732111e+00,\n        -6.17750780e+00,  5.54877909e+00],\n       [-3.63749982e+01,  5.55591616e+00,  5.97446684e+00,\n         3.02279596e+00, -3.30542437e+00],\n       [ 3.19043362e+01, -9.29263336e+00,  1.85181690e+00,\n         1.01349628e+01, -2.37058592e+00],\n       [ 3.62765262e+01,  1.96671011e+00,  1.09120015e+00,\n         6.63487027e+00, -8.59972398e-01],\n       [ 1.77999268e+01,  4.32134130e+00,  4.53765279e+00,\n        -1.34673088e+01, -1.16875727e+00],\n       [-5.25996690e+00,  1.33213010e+01,  1.74140499e+00,\n         4.71751635e+00,  8.76779964e+00],\n       [ 7.95049466e+00, -3.71483964e+00,  2.12290605e-01,\n         3.62280679e+00, -2.44456126e+00],\n       [ 4.88572471e-01, -4.42148741e+00,  5.45577841e+00,\n        -1.29349705e+01, -2.03687023e+00],\n       [-1.86892777e+01,  7.04801042e-01, -9.17634055e+00,\n        -1.16258207e+01, -9.59454518e+00],\n       [ 2.41722949e+01, -8.90883920e+00,  1.81260470e+00,\n         2.75578810e+00,  4.00236343e+00],\n       [ 3.66661965e+01,  5.52252755e-01,  2.94042439e+00,\n         1.26737840e+00, -4.93647434e+00],\n       [ 9.50461849e+00, -1.58222347e+01,  8.23173689e+00,\n         4.28177479e+00, -5.13712979e+00],\n       [ 2.75754139e+01,  1.62357398e+01,  1.65729689e+00,\n        -1.41684221e+00,  6.17813086e+00],\n       [-1.51693257e+01, -6.20554707e+00,  5.03081974e+00,\n         1.12129146e-01, -4.09598561e+00],\n       [ 2.74311655e+01,  1.36360397e+01,  4.61297245e+00,\n         8.61678055e+00,  3.04585562e+00],\n       [-5.47241824e+00,  1.72969967e+01,  1.28083986e+01,\n        -6.77133066e-01, -1.35979233e+00],\n       [ 2.20846466e+01,  8.40542611e+00,  1.22435299e+01,\n         5.20309150e+00, -2.16727571e+00],\n       [ 1.66932608e+01, -8.03098568e+00,  2.02068249e+00,\n         1.70874550e+00, -2.41276147e+00],\n       [-1.32500937e+01, -1.12800290e+00,  4.53962048e+00,\n        -1.42748280e+00, -1.66217955e+00],\n       [-3.50583690e+01,  1.11593437e+00,  2.18127401e+00,\n         6.64976960e+00,  6.95800413e-01],\n       [-2.84664240e+01,  4.66642477e+00,  6.17183086e+00,\n        -1.08166647e+01,  2.08842742e+00],\n       [ 2.78231830e+00,  3.87935737e+00,  7.99356252e+00,\n        -7.73388773e+00,  2.36120839e+00],\n       [ 2.50600466e+01,  1.05252315e+01,  4.69574777e+00,\n         2.81307266e+00,  4.46387918e+00],\n       [-8.21830063e+00, -8.35578349e+00, -3.75205043e+00,\n         1.86998322e+00,  2.05469519e+00],\n       [-2.28916931e+01, -2.31308992e+00,  6.26909400e+00,\n        -4.48911994e+00,  4.36031393e+00],\n       [ 2.87234425e+01, -4.51097103e+00, -2.90170730e-01,\n        -1.68002540e+00, -1.15812082e+01],\n       [ 9.46215003e+00, -7.62310031e+00,  6.33629311e+00,\n         3.07508479e+00, -8.97426191e+00],\n       [-1.27408087e+01,  4.69970994e-01,  1.90663150e+00,\n         3.26827018e+00,  4.43735987e+00],\n       [ 1.26854720e+01,  1.73520280e+01, -9.04859156e+00,\n        -5.49041886e+00, -7.63611744e+00],\n       [-2.58321366e+01,  1.07863972e+01,  4.14238334e+00,\n         5.26676491e+00, -2.99203558e+00],\n       [-1.48881798e+01,  2.46392952e+00, -2.21868658e+00,\n        -6.44506252e+00,  1.22505439e+01],\n       [ 1.77499497e+01,  2.07628188e+00,  1.02105448e+01,\n         1.06776362e+00,  8.55272292e+00],\n       [-2.12595276e+01, -1.51069017e+00, -4.48537860e+00,\n        -3.44060547e+00, -6.68206474e+00],\n       [ 3.58927317e+00, -1.52322210e+00,  9.59040211e+00,\n        -4.58902771e+00, -5.23621997e+00],\n       [ 1.87503087e+01, -1.29329318e+01, -6.94110522e+00,\n        -8.25703736e+00, -2.74292990e+00],\n       [ 3.34049240e+01,  4.24748127e+00,  1.37052930e+01,\n         8.22930940e+00,  5.13122196e+00],\n       [-8.92256911e+00,  5.23666768e+00,  4.59775213e+00,\n        -2.32282953e+00, -7.40312673e+00],\n       [-4.34746647e+00,  4.56016620e+00, -8.93770613e+00,\n         1.41817005e+01, -2.71242961e+00],\n       [-5.13154765e+00,  1.33949331e+01,  8.97549962e+00,\n        -6.86198714e+00,  7.16746821e+00],\n       [-2.81980846e+01,  7.01652856e+00,  2.78468417e+00,\n        -1.39679927e+00,  6.05734060e-01],\n       [ 1.58341023e+01, -1.67305717e+00, -1.75063848e+00,\n         1.20056202e+00, -8.68631543e+00],\n       [ 1.32565326e+01, -3.31574916e+00,  2.03707782e+00,\n        -3.71985235e+00, -3.10953990e+00],\n       [-1.04881367e+01,  8.80087328e+00, -9.20898656e+00,\n        -1.75247827e+00, -3.29288309e+00],\n       [-3.02416591e+01,  1.07617563e+01,  1.23274951e+00,\n         7.56914196e+00,  5.84078910e-01],\n       [-3.43081924e+01,  2.88913442e-01, -1.73270470e+00,\n        -2.34224043e+00, -7.52477121e+00],\n       [ 2.88009403e+01, -2.62943855e+00, -4.95791648e+00,\n        -1.44655837e+00, -1.15952543e+00],\n       [-2.44606249e+01,  2.85139830e+00, -1.30332056e+00,\n         4.93843821e+00, -3.61754291e+00],\n       [-2.92881868e+01,  3.12609735e+00,  5.02202299e+00,\n        -3.87499853e+00, -9.84569588e-01],\n       [-1.16525673e+01,  1.74631508e+00, -7.15970439e+00,\n        -1.12145317e+01, -5.96632592e+00],\n       [ 2.04177233e+01, -3.15920766e+00,  8.65199407e-01,\n         1.19312853e+01,  1.40282447e+00],\n       [-4.00602852e+00,  1.15473230e+00,  6.91992455e+00,\n         5.61964383e+00,  5.19409774e+00],\n       [-5.93443654e+00,  6.02651011e+00, -5.17269453e+00,\n        -4.75566942e+00,  1.92783651e+00],\n       [-4.99355994e+00, -5.12934783e+00,  3.58593349e-01,\n         1.41043362e+01, -6.86799356e+00],\n       [-1.46523572e+01, -4.57538267e-01, -2.87418424e+00,\n         1.18571867e+01, -2.03257237e+00],\n       [ 3.07339086e+01,  1.13376349e+01,  1.71086630e+00,\n        -7.92190135e+00, -3.34624920e+00],\n       [ 3.93595897e+00, -5.95159544e+00, -3.30991261e+00,\n        -5.10063566e+00,  1.21276878e+00],\n       [-3.26074133e+01, -6.11376176e+00,  3.24752698e+00,\n         6.39390513e+00, -8.92263847e-01],\n       [-2.24023483e+00, -7.22932888e-02,  4.73604303e+00,\n        -1.32414596e+01,  8.78099308e+00],\n       [ 1.62841588e+01, -9.76013475e+00, -2.36064437e+00,\n        -3.02434719e-01,  3.46956869e+00],\n       [-1.45605285e+01, -1.23012501e+01,  4.34323736e+00,\n         6.71181384e+00,  3.54882720e+00],\n       [-1.98371833e+01,  2.94715826e+00,  2.25937469e-01,\n         3.45849853e+00,  8.75524036e+00],\n       [-2.31448581e+01, -6.58276550e+00,  7.68436942e+00,\n         3.29016312e+00,  7.99150224e+00],\n       [ 2.78175236e+01, -1.46111168e+00, -3.55131601e+00,\n        -4.58529833e+00,  1.64006212e+00],\n       [ 4.62099793e+00,  6.55008762e+00, -7.51458853e+00,\n        -1.31653265e+00, -4.55263728e+00],\n       [-2.90888809e+01, -4.06337855e+00,  2.71015719e+00,\n         4.92714531e+00,  7.04345967e+00],\n       [-1.08311429e+01,  3.34425200e+00, -3.66034430e+00,\n         1.62251944e+00, -7.28407221e+00],\n       [ 1.94275934e+01, -9.97998784e+00, -1.79143238e+00,\n        -2.18943023e+00,  1.19234908e+00],\n       [-1.69889614e+00, -2.16707406e-01,  3.76814357e+00,\n         2.71978490e+00,  7.05242716e+00],\n       [ 3.32728567e+00, -5.57228466e+00,  8.14818126e-02,\n        -4.47433137e+00,  1.98207169e+00],\n       [ 5.08048361e+00, -1.03195093e+01, -3.60385302e-01,\n         6.96871129e+00, -1.65821454e+00],\n       [ 1.03412380e+01,  2.62056258e+00,  1.33338480e+00,\n         1.82494079e+00, -3.25564115e+00],\n       [ 1.25753469e+01, -8.67129472e+00, -1.07582333e+01,\n         9.56163462e+00,  9.48951703e+00],\n       [-1.12457322e+01, -2.22908600e+00, -7.32849965e-01,\n        -3.65425410e+00, -1.77485655e+00],\n       [ 1.09717690e+01, -8.46026646e+00,  8.71063540e+00,\n        -2.49433730e+00, -1.60105096e+00],\n       [ 1.19113486e+01,  6.95350575e+00, -9.74343014e+00,\n         1.79808098e+00, -4.69223625e+00],\n       [ 6.35355013e+00,  2.99510780e+00,  5.82413302e-01,\n         6.09567701e+00, -2.86204847e+00],\n       [ 4.35468263e+01,  3.87421471e+00,  1.31488784e-01,\n         9.56010168e+00, -6.98299665e-01],\n       [-6.41354541e+00, -3.65377814e+00, -1.58613190e+00,\n         3.14952723e+00,  3.57310500e+00],\n       [-1.83071512e+01,  2.67943240e+00, -7.33703842e+00,\n         1.14969873e+01, -6.00183994e+00],\n       [-1.14967710e+01,  1.19634160e+00,  2.05056704e+00,\n         9.64640181e+00,  1.09353192e+01],\n       [ 1.19388883e+01, -2.21511698e-01, -7.66212803e+00,\n        -7.36119191e-01,  8.54328202e-01],\n       [-8.36745242e+00, -1.57527368e+01,  2.37688116e+00,\n         2.31755360e+00,  6.79797604e+00],\n       [-3.18705853e+01,  1.28750482e+00,  7.80008431e+00,\n        -1.96742133e+00,  7.47677308e+00],\n       [-2.90282796e-01,  7.95119891e+00, -8.90775571e+00,\n        -2.02404252e+00,  1.88114114e-01],\n       [ 4.02027481e+00, -7.16484171e+00, -2.91142921e+00,\n         3.50831453e+00, -5.58782765e+00],\n       [ 2.93293387e+01, -1.99243185e+00,  8.59832470e-01,\n        -6.37100920e+00,  4.60403239e+00],\n       [ 2.56981194e+00, -1.17813065e+01,  4.06779343e+00,\n        -1.43837910e-01, -1.05609021e+00],\n       [ 2.27229949e+01, -7.14473449e+00, -9.11557864e+00,\n        -7.92957478e+00, -8.73283554e+00],\n       [ 6.54574020e+00, -2.27567523e+00, -7.09510478e+00,\n        -3.02614488e+00, -3.03522613e+00],\n       [ 1.35368989e+01, -6.16348151e+00,  5.05265674e+00,\n         5.74538174e+00, -5.58490706e+00],\n       [ 1.14745144e+01,  2.17973250e+00, -7.63390038e+00,\n         5.42164029e+00,  5.86427676e+00],\n       [-1.16971089e+01, -8.86475008e+00, -3.63714396e+00,\n        -1.38379698e+00, -4.80377090e+00],\n       [ 4.55426837e+00, -1.15169002e+01, -9.56865897e-01,\n        -7.35846277e+00,  1.11047591e+00],\n       [ 1.08759892e+01,  1.43465719e+01, -1.04959316e+01,\n        -1.16009535e+00,  1.60671842e+01],\n       [ 1.24429155e+01, -6.48004343e+00, -4.03048524e+00,\n        -2.28161003e+00,  2.58619593e+00],\n       [ 2.07945770e+01,  6.30017525e-01,  4.95995846e+00,\n         4.78379367e+00,  5.99042792e-01],\n       [ 1.92814303e+01, -4.68765599e+00, -7.07524200e+00,\n         9.05476173e+00,  7.40679699e+00],\n       [ 2.05385809e+01, -2.41467304e+00,  8.88148407e+00,\n        -4.65253394e-01, -7.17569219e-01],\n       [-1.45863823e+01,  3.78201341e+00,  6.79461149e+00,\n         3.41594827e+00,  4.73766051e-01],\n       [ 9.51331112e+00, -4.16730759e+00, -8.34955278e+00,\n        -7.66699692e+00,  5.65408354e+00],\n       [-9.85433997e-01, -7.24467998e+00, -2.86632021e+00,\n        -7.06408344e+00, -2.40052919e+00],\n       [-1.66831422e+01, -6.29776099e+00, -2.79565665e+00,\n        -3.99269089e+00,  9.88416524e-02],\n       [-2.39368143e+01, -9.59356022e+00,  6.57033610e+00,\n         2.14853164e+00,  3.91270145e-01],\n       [-4.68288895e+00, -6.12821649e+00, -6.52459435e-01,\n        -5.53043347e+00,  1.38184226e-02],\n       [ 1.89801457e+00, -1.08900615e+01, -8.39070060e+00,\n        -2.94585171e+00, -4.61825468e+00],\n       [ 2.48991324e+01,  6.19169018e-01,  3.71209663e+00,\n        -6.29206979e-01, -5.38837443e-01],\n       [-2.91110201e+01, -1.55973824e+00, -1.85580884e+00,\n         3.62583869e+00, -3.17915901e-01],\n       [-6.06012752e+00, -6.90453493e+00, -9.34705855e+00,\n        -3.68284469e+00, -1.56444184e+00],\n       [-2.28992384e+00, -4.84517863e+00,  5.92365817e+00,\n        -1.70190259e+00,  1.66444266e+00],\n       [ 5.06787447e+00,  2.14073139e+01, -1.21295425e+01,\n        -1.94649242e+00,  6.73312719e+00],\n       [-2.24069761e+01,  3.24436337e+00, -4.36511644e+00,\n         4.69869305e+00, -9.29169776e+00],\n       [-5.95136767e+00, -6.36086833e+00, -3.65116574e+00,\n        -6.79194347e+00,  2.97239712e+00],\n       [ 2.82464607e+01,  6.29180525e+00, -8.64648706e-02,\n         4.15139708e+00,  6.96671696e+00],\n       [-9.68349974e+00, -1.72848964e+00, -5.15912819e+00,\n        -8.62359876e+00, -1.92488233e+00],\n       [ 2.15060895e+01,  5.18870552e+00, -1.02483228e+01,\n        -2.92067465e+00,  7.23692366e+00],\n       [-6.57279467e+00,  1.26972256e+01, -6.30689962e+00,\n         1.01862018e+01,  2.66818605e+00],\n       [ 2.11284368e+00,  1.17493416e+01, -4.52560027e+00,\n         5.75207793e+00,  1.09643732e+01],\n       [ 1.16485222e+01,  3.08938055e+00, -1.02468874e+01,\n        -2.51350542e+00,  4.03451642e+00],\n       [ 9.21449431e+00, -1.26094458e+01,  3.96154180e+00,\n        -9.78420576e+00,  1.22741578e+00],\n       [ 3.18342319e+01, -4.09762928e+00,  5.78311319e+00,\n         7.07288934e+00, -3.27149250e-01],\n       [-1.36299212e+01,  8.83563969e+00, -8.28856251e-01,\n         2.76842935e+00,  3.76284628e-01],\n       [ 1.71507080e+01, -1.12921583e+01, -1.76841151e+00,\n         3.97856967e+00,  1.41426870e+00],\n       [-2.08477163e+01, -5.82746358e+00,  1.87952305e+00,\n         5.36663557e+00, -3.06040614e+00],\n       [-2.67095464e+01, -9.96579444e+00, -1.62170872e+00,\n         3.48976776e+00,  9.05940833e-01],\n       [ 1.02005462e+01,  8.64022146e+00, -1.05174225e+01,\n        -9.91827010e-02, -4.52056321e+00],\n       [ 9.75027450e+00,  1.41164994e+01, -2.31083644e+00,\n        -3.93975961e+00,  7.46732858e+00],\n       [-9.29096061e+00,  4.32672500e+00, -1.02831088e+01,\n         3.44668794e+00,  3.62756210e-01],\n       [-4.65051582e+00, -1.11672351e+00, -2.41307796e+00,\n         3.25908036e+00,  1.11730549e+01],\n       [ 1.37338848e+01, -7.94241114e+00, -1.99982730e+00,\n         3.53782539e+00, -1.56175284e+00],\n       [-1.65858633e+01, -6.53255864e-02, -1.04150256e+01,\n        -5.99735926e+00, -1.72834379e-01],\n       [-1.07606237e+01, -7.84433863e+00, -6.93046682e+00,\n         2.31363920e+00,  3.04254964e+00],\n       [-1.83689058e+01,  7.59636328e+00, -1.27859039e+01,\n        -1.88684616e+00, -6.05517243e+00],\n       [-2.75788756e+01, -7.78655254e+00, -1.15182611e-01,\n         1.74551120e+00, -8.78189080e-01],\n       [-1.23679131e+00,  2.21482797e+00, -1.40819352e+00,\n         1.11740318e+00, -8.54026018e-01],\n       [ 2.27265416e+01,  4.37759871e+00,  2.24609468e+00,\n         3.20136068e+00,  5.73265720e+00],\n       [-1.59410574e+01, -3.82795013e+00,  2.43267265e-01,\n         1.48237350e+00,  5.01188102e+00],\n       [-3.11916611e+00, -9.18964731e+00,  5.58345867e+00,\n        -6.62787717e+00, -8.23793018e+00],\n       [ 1.89933076e+01, -8.51332981e-01, -7.93010918e+00,\n         8.79312072e-01,  6.61254503e+00],\n       [ 1.67713405e+01,  3.43481221e+00, -1.39762654e+01,\n        -5.37196620e+00,  3.28728775e+00],\n       [-1.79297286e+00, -8.89036206e-01, -3.27594642e-01,\n        -3.88848441e+00, -3.05177141e+00],\n       [ 1.56619674e+01, -1.34089220e+01, -1.12831735e+01,\n         4.00174691e+00,  5.50959894e+00],\n       [-6.65097016e+00, -3.37562201e+00, -1.27301064e+01,\n         4.37987124e+00, -3.22132981e+00],\n       [-1.15909970e+00, -9.00928901e+00, -2.70346264e+00,\n         6.83919755e+00,  4.49354438e+00],\n       [ 2.38621570e+01, -8.28526814e+00, -9.04416458e-01,\n         5.05020797e-01,  4.26085161e+00],\n       [-3.51796897e-01,  4.54862105e+00,  1.17932127e+01,\n         2.75841961e-01, -5.52144959e+00],\n       [-1.42051083e+01, -2.14051145e+00, -4.75335975e+00,\n        -1.82517421e+00, -9.23847914e-01],\n       [-2.01694962e+01, -3.01896312e+00, -2.64038366e+00,\n         8.37040244e+00,  5.72797768e-01]])"
          },
          "execution_count": 89,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "pca.components_",
      "execution_count": 91,
      "outputs": [
        {
          "data": {
            "text/plain": "array([[-0.48424397, -0.42202931, -0.42461477, -0.42959103, -0.47175511],\n       [ 0.07210494, -0.08149798,  0.24803223,  0.58940457, -0.76107899],\n       [ 0.61037107, -0.78014061, -0.05990336,  0.00204942,  0.12343085],\n       [ 0.3010751 ,  0.13846051,  0.56838866, -0.67857473, -0.32657791],\n       [ 0.54507761,  0.43296496, -0.65690808, -0.0870973 , -0.27625636]])"
          },
          "execution_count": 91,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "$$performance = 0.484 \\times read + 0.422 \\times write + 0.42 \\times matn + 0.49 \\times science + 0.47 \\times socst$$"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "dat = pd.read_csv(\"~/dl/r25h_os26c.csv\", encoding='cp1251')",
      "execution_count": 93,
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": "/usr/local/lib/python3.6/site-packages/IPython/core/interactiveshell.py:2728: DtypeWarning: Columns (114,128,132,133,134,136,137,140,142,143,145,147,148,149,150,151,152,153,154,155,156,158,159,161,163,164,165,166,167,168,169,170,171,172,173,175,176,178,180,181,182,183,184,185,186,187,188,189,190,191,193,194,196,198,199,200,201,202,203,204,205,206,207,208,209,210,212,213,215,217,218,219,220,222,223,224,225,226,227,228,229,230,232,233,235,237,238,239,240,241,242,243,244,245,246,247,248,249,250,251,253,254,256,258,259,260,262,263,264,265,266,267,268,269,270,271,272,273,275,276,278,280,281,282,285,286,287,288,289,290,291,292,293,294,295,296,301,303,304,305,308,309,310,311,312,313,314,315,316,317,318,319,618,1007,1008) have mixed types. Specify dtype option on import or set low_memory=False.\n  interactivity=interactivity, compiler=compiler, result=result)\n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from scipy.stats import chi2_contingency",
      "execution_count": 96,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "contigency_matrix = np.array([[45, 67], \n                              [30, 15]])",
      "execution_count": 98,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res = chi2_contingency(contigency_matrix)",
      "execution_count": 100,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "res",
      "execution_count": 101,
      "outputs": [
        {
          "data": {
            "text/plain": "(7.9968506662149945,\n 0.004685878096189338,\n 1,\n array([[53.50318471, 58.49681529],\n        [21.49681529, 23.50318471]]))"
          },
          "execution_count": 101,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.6.4",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "gist": {
      "id": "",
      "data": {
        "description": "Stats Notes.ipynb",
        "public": true
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}