sliderule_dsi_inferential_statistics_exercise_1-Copy2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# What is the True Normal Human Body Temperature? \n",
    "\n",
    "#### Background\n",
    "\n",
    "The mean normal body temperature was held to be 37$^{\\circ}$C or 98.6$^{\\circ}$F for more than 120 years since it was first conceptualized and reported by Carl Wunderlich in a famous 1868 book. But, is this value statistically correct?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>Exercises</h3>\n",
    "\n",
    "<p>In this exercise, you will analyze a dataset of human body temperatures and employ the concepts of hypothesis testing, confidence intervals, and statistical significance.</p>\n",
    "\n",
    "<p>Answer the following questions <b>in this notebook below and submit to your Github account</b>.</p> \n",
    "\n",
    "<ol>\n",
    "<li>  Is the distribution of body temperatures normal? \n",
    "    <ul>\n",
    "    <li> Although this is not a requirement for the Central Limit Theorem to hold (read the introduction on Wikipedia's page about the CLT carefully: https://en.wikipedia.org/wiki/Central_limit_theorem), it gives us some peace of mind that the population may also be normally distributed if we assume that this sample is representative of the population.\n",
    "    <li> Think about the way you're going to check for the normality of the distribution. Graphical methods are usually used first, but there are also other ways: https://en.wikipedia.org/wiki/Normality_test\n",
    "    </ul>\n",
    "<li>  Is the sample size large? Are the observations independent?\n",
    "    <ul>\n",
    "    <li> Remember that this is a condition for the Central Limit Theorem, and hence the statistical tests we are using, to apply.\n",
    "    </ul>\n",
    "<li>  Is the true population mean really 98.6 degrees F?\n",
    "    <ul>\n",
    "    <li> First, try a bootstrap hypothesis test.\n",
    "    <li> Now, let's try frequentist statistical testing. Would you use a one-sample or two-sample test? Why?\n",
    "    <li> In this situation, is it appropriate to use the $t$ or $z$ statistic? \n",
    "    <li> Now try using the other test. How is the result be different? Why?\n",
    "    </ul>\n",
    "<li>  Draw a small sample of size 10 from the data and repeat both frequentist tests. \n",
    "    <ul>\n",
    "    <li> Which one is the correct one to use? \n",
    "    <li> What do you notice? What does this tell you about the difference in application of the $t$ and $z$ statistic?\n",
    "    </ul>\n",
    "<li>  At what temperature should we consider someone's temperature to be \"abnormal\"?\n",
    "    <ul>\n",
    "    <li> As in the previous example, try calculating everything using the boostrap approach, as well as the frequentist approach.\n",
    "    <li> Start by computing the margin of error and confidence interval. When calculating the confidence interval, keep in mind that you should use the appropriate formula for one draw, and not N draws.\n",
    "    </ul>\n",
    "<li>  Is there a significant difference between males and females in normal temperature?\n",
    "    <ul>\n",
    "    <li> What testing approach did you use and why?\n",
    "    <li> Write a story with your conclusion in the context of the original problem.\n",
    "    </ul>\n",
    "</ol>\n",
    "\n",
    "You can include written notes in notebook cells using Markdown: \n",
    "   - In the control panel at the top, choose Cell > Cell Type > Markdown\n",
    "   - Markdown syntax: http://nestacms.com/docs/creating-content/markdown-cheat-sheet\n",
    "\n",
    "#### Resources\n",
    "\n",
    "+ Information and data sources: http://www.amstat.org/publications/jse/datasets/normtemp.txt, http://www.amstat.org/publications/jse/jse_data_archive.htm\n",
    "+ Markdown syntax: http://nestacms.com/docs/creating-content/markdown-cheat-sheet\n",
    "\n",
    "****"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>temperature</th>\n",
       "      <th>gender</th>\n",
       "      <th>heart_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>99.3</td>\n",
       "      <td>F</td>\n",
       "      <td>68.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>98.4</td>\n",
       "      <td>F</td>\n",
       "      <td>81.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>97.8</td>\n",
       "      <td>M</td>\n",
       "      <td>73.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>99.2</td>\n",
       "      <td>F</td>\n",
       "      <td>66.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>98.0</td>\n",
       "      <td>F</td>\n",
       "      <td>73.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   temperature gender  heart_rate\n",
       "0         99.3      F        68.0\n",
       "1         98.4      F        81.0\n",
       "2         97.8      M        73.0\n",
       "3         99.2      F        66.0\n",
       "4         98.0      F        73.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import scipy.stats as stats \n",
    "\n",
    "df = pd.read_csv('data/human_body_temperature.csv')\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>temperature</th>\n",
       "      <th>heart_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>130.000000</td>\n",
       "      <td>130.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>98.249231</td>\n",
       "      <td>73.761538</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.733183</td>\n",
       "      <td>7.062077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>96.300000</td>\n",
       "      <td>57.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>97.800000</td>\n",
       "      <td>69.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>98.300000</td>\n",
       "      <td>74.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>98.700000</td>\n",
       "      <td>79.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>100.800000</td>\n",
       "      <td>89.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       temperature  heart_rate\n",
       "count   130.000000  130.000000\n",
       "mean     98.249231   73.761538\n",
       "std       0.733183    7.062077\n",
       "min      96.300000   57.000000\n",
       "25%      97.800000   69.000000\n",
       "50%      98.300000   74.000000\n",
       "75%      98.700000   79.000000\n",
       "max     100.800000   89.000000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x22469148ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x22469beabe0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt \n",
    "df.temperature.hist()\n",
    "plt.xlabel('Temperature')\n",
    "plt.ylabel('Frequency')\n",
    "plt.show()\n",
    "\n",
    "df.heart_rate.hist()\n",
    "plt.xlabel('Heart Rate')\n",
    "plt.ylabel('Frequency')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gender\n",
       "F    65\n",
       "M    65\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('gender').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NormaltestResult(statistic=2.703801433319203, pvalue=0.2587479863488254)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import scipy.stats as stats\n",
    "stats.normaltest(df.temperature)\n",
    "# since the value of p is 0.258, we cannot reject the null hypothesis that is that the data is normal.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEKCAYAAAAIO8L1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xd8VHX2//HXCYiKBaVYVk2iyCJWVKz7sy2uq7KLigVZ7CUWVFisGFZdpSNYQTeK2KIL9q6LfhXWvkFRQewkiA0BwRJKCOf3x+cGJmEmmYRMJpm8n4/HPGbmM3funIySk3vP/ZyPuTsiIiJVZaU7ABERaZyUIEREJC4lCBERiUsJQkRE4lKCEBGRuJQgREQkLiUIERGJK2UJwszuMbP5ZjYzZqytmU0xs8+j+82j8Z3M7C0zW25ml6UqJhERSV4qjyDuBY6sMnYV8Iq7dwJeiZ4DLAIuAW5MYTwiIlILLVO1Y3efZma5VYaPAQ6NHt8HvAZc6e7zgflm1qM2n9G+fXvPza36ESIiUp3p06cvcPcONW2XsgSRwJbu/h2Au39nZlvUdgdmlgfkAWRnZ1NUVFTPIYqIZDYzK0lmuyZXpHb3Anfv5u7dOnSoMQGKiEgdNXSC+MHMtgaI7uc38OeLiEiSGjpBPA2cHj0+HXiqgT9fRESSlLIahJk9TChItzezecC1wAhgspmdDcwFToy23QooAjYFVpnZAGBnd/85VfGJiEj1UnkVU58EL3WPs+33wLapikVERGqvyRWpRUSkeoWFkJsLWVnhvrCwbvtp6MtcRUQkhQoLIS8PSkvD85KS8Bygb9/a7UtHECIiGSQ/f01yqFBaGsZrSwlCRCSDzJ1bu/HqKEGIiGSQ7OzajVdHCUJEJIMMHQqtW1cea906jNeWEoSISAbp2xcKCiAnB8zCfUFB7QvUoKuYREQyTt++dUsIVekIQkRE4lKCEBGRuJQgREQkLiUIERGJSwlCRETiUoIQEZG4lCBERCQuJQgREYlLCUJEROJSghARkbiUIEREJC4lCBERiUsJQkRE4lKCEBGRuJQgREQkLiUIERGJSwlCRETiUoIQEZG4lCBERCQuJQgREYlLCUJEROJSghARkbiUIEREJC4lCBERiUsJQkRE4lKCEBGRuJQgREQkrpQlCDO7x8zmm9nMmLG2ZjbFzD6P7jePxs3MbjWzL8zsQzPbK1VxiYg0VoWFkJsLWVnhvrAwvfGk8gjiXuDIKmNXAa+4eyfgleg5wFFAp+iWB9yRwrhERBqdwkLIy4OSEnAP93l5dUwS5eUweTIsW7ZOMaUsQbj7NGBRleFjgPuix/cBx8aM3+/B28BmZrZ1qmITEWls8vOhtLTyWGlpGE9aWRlMnAhdukDv3iFJrIOGrkFs6e7fAUT3W0Tj2wBfx2w3Lxpbi5nlmVmRmRX9+OOPKQ1WRKShzJ1bu/FKli2D8eNhxx3hrLNg443hscfglFPWKabGUqS2OGMeb0N3L3D3bu7erUOHDikOS0SkYWRn124cgN9+g7FjYfvtoV8/2GYbeO45mD4devUKxYx10NAJ4oeKU0fR/fxofB6wXcx22wLfNnBsIiL1LtnC89Ch0Lp15bHWrcP4WpYsCS/k5MCll8LOO8P//R+88QYcfTRYvL+5a6+hE8TTwOnR49OBp2LGT4uuZtofWFJxKkpEpKmqTeG5b18oKAi/883CfUFBGF9twQL4xz/Ci4MHw/77w5tvwiuvwGGH1VtiqGDucc/krPuOzR4GDgXaAz8A1wJPApOBbGAucKK7LzIzA24nXPVUCpzp7kU1fUa3bt28qKjGzURE0iI3NySFqnJyoLi4Fjv67jsYMwbuvDOcVjr++FC93nPPOsVlZtPdvVtN27Ws096T4O59ErzUPc62DvRLVSwiIumwToXnig1HjYK77w5XKPXpA4MGwS671FuM1UlZghARae6ys+MfQVRbeAb44gsYMQLuuy+cNjr9dLjyynCVUgNqLFcxiYhknFoVngFmzQpFh86d4cEH4fzzQ7K4664GTw6gBCEikjJJFZ4B3nsv1BV23RWeeipcmVRcDLfdlsThRuroFJOISAr17RsnIVR4800YMgReeAHatAlXKPXvD+3aNWiMiShBiIg0JHd49dWQGF59Fdq3h2HD4MILQ5JoRJQgREQagjs8/3xIDG+/DVtvHWZB5+XBRhulO7q4lCBERFJp1Sp44omQGGbMCIWI8ePhzDNhgw3SHV21VKQWEUmFlSvDlUi77gonnBAmuE2cCJ9/Dhdc0OiTAyhBiIjUrxUrwsS2zp3h1FOhRQt4+GGYPRvOOAPWWy/dESZNCUJEpD4sXQq33w4dO8K550LbtvDkk/DBB3DyySFRNDGqQYiIrItffgk9ksaMgR9+gP/3/8IRxBFH1HvzvIamBCEiUheLF4eJbDffDIsWwZ/+FDqsHnxwuiOrN0oQIiK18eOPcNNN4XTSL7/AX/8aOqvut1+6I6t3qkGIiCTj229h4MDQw3vECDjqKJgxg8LeT5Pbe78aFwRqinQEISJSneJiGDkS7rkHystD34xBg2CnnVYvCFRaGjatWBAIqmmv0YTUeARhZh3NbP3o8aFmdomZbZb60ERE0uizz8Jktk6dYMKE8Pizz0IL7p12AsKZpYrkUKG0NIxngmROMT0GlJvZjsAEYHvgoZRGJSKSLh99FC5L3WknmDQJ+vWDr74KVyrtsEOlTdd5QaBGLpkEscrdVwLHATe7+9+BrVMblohIA/vf/+DYY2H33eG55+CKK2DOnHCV0rbbxn1Lok7caezQXa+SSRBlZtYHOB14NhprOlMBRUSq89//wpFHwr77wtSpfNjrOvbYrISsUSPI3W9LCgtD4Tk3l7UK0bVeEKiJSaZIfSZwPjDU3eeY2fbAg6kNS0Qkhdzh5ZdDA71p06BDBxgxgkltL+CsAZtWKjqfeWaY77ZixZqxqoXo/PxwWik7OySHTChQA5i717yR2YZAtrt/mvqQktetWzcvKipKdxgi0lS4w7PPhsTw7ruwzTbhVNI550Dr1uTmxl9DOp6cnHCBU1NkZtPdvVtN2yVzFdNfgRnAi9Hzrmb29LqHKCLSQMrLYfJk6NoVevYMk93+9S/48ku45JLV54lqU1zOlEJ0dZKpQVwH7AssBnD3GYQrmUREGreyMrj/fthlF+jdmy8/Xs5p3E/H8s8o3CiPwkfXr1RbaNs2+V1nSiG6OsnUIFa6+xKr3HSq5vNSIiLpsnw53HtvmPFcXMxP2btzcavJPLyiF6toAXPj1xbWWw9atVozBmEsdjvIrEJ0dZI5gphpZn8DWphZJzO7DXgzxXGJiNReaSnccktouX3++bDFFvD00+zJDApXnBiSQ6SsrPIv/YqxTTYJ9QWzcD9xYphEHTtWUJA5hejqJHMEcTGQDywHHgZeAm5IZVAiIrXy889hGc+xY0N94ZBDwhFE9+5gxtxjkt/VokWwYMHa480hIVRVY4Jw91JCgsiQyeMikjEWLYJbbw1HDYsXw5//HK45PeigSptlZyd/dVJzqC0kK2GCMLNnqKbW4O49UxKRiEhNfvgBbrqJslvGsd6yX3mCYxnXJp8ZRd1YdEj4JX/00fD88+Fqo7ZtVVuoi+qOIG5ssChERJIxbx6MHg0FBaxavoIns07ieq5mJrvBkjWblZTAHXeseb5wYUgI7dqFg46KCW2QuZPc6kPCBOHuUxsyEBGRhL76KrTcnjgxTHY79VS6v3QVr337+6R3UVYGG2+8dn1BCSGx6k4xTXb3k8zsI+KcanL33VMamYjI7NkwfDg89BC0aBFmPF9xBeTmMrUOy501h8lt9am6U0z9o/u/NEQgIiKrzZhByXnD2O7dR1nKhty3QX/Gb3gpH9/5O9pODpsk0SVoLSpA107CHOzu30UPL3T3ktgbcGHDhCcizco774Q1nvfck83ffZHhDCKXYvotG8Osn36He6gnLFxY+12rAF17yRyk/SnO2FH1HYiINFPuMHUq/OlPsP/+8OabjGlzPTmUMJihLKBDUrtp1y7cKiazXXBB85zcVp+qq0FcQDhS2MHMPox5aRPgjVQHJiIZzh1eein8Wf/667DlluEKpfPP5/JNN65VPx+z+JPbZN1UdwTxEPBX4OnovuK2t7ufsi4famb9zWymmc0yswHR2B5m9paZfWRmz5jZpuvyGSLSuKxedMdW0avFkxRl7QNHHcXXrxdzZevb2LZsDnb5ZbTcbONa1xdUW0iN6moQS9y92N37APOAMsLVTBubWZ3/c5jZrsC5hA6xewB/MbNOwN3AVe6+G/AEcHldP0NEGpfCQjj/3HIOKHmYD9iDx1cdx2Ys5mzupiNfMqr0Ir5ZtCEQOnPXhmoLqVNjqw0zu4jQ8vsHYFU07EBdL3PtArwdtfDAzKYS1rvuDEyLtplC6Pn0jzp+hog0FmVlzOj/INOXDuf3fM7HdKEvDzKJ3pQn1Q4u1BYgTHKraMkdO+FNtYXUSOa/zgCgs7vX4bqBuGYCQ82sHbAUOBooisZ7Ak8BJwLbxXuzmeUBeQDZOq4UabyWLQsT20aOZPTCEt5jT3rxGE9yLJ7U9TGB6gvpk8x/pa+pNIl93bj7bGAk4SjhReADYCVwFtDPzKYTCuErEry/wN27uXu3Dh2Su7pBRBrQb7+Frqo77AAXXghbb80ZWzzH3kznCXrVKjmA6gvplMwRxFfAa2b2HKHlNwDuPrauH+ruE4AJAGY2DJjn7p8AR0Rjvwd61HX/IpIGS5bAuHFw002wYAFvrn8Y/+BBPvjsMJYtt5rfH4fqC+mVTIKYG91aRbd1ZmZbuPv8qNjdCzggZiwLGAzcWR+fJSIptmBBaLd9222wZAnf7HE0p/6Sz6vLDwyvL1r7LVlZsGpV6J5RXr6mxrBw4ZqxnBzVF9ItmfUg/pmCz30sqkGUAf3c/afo0td+0euPAxNT8LkiUl++/x7GjAltU3/7DXr1gvx8/tBrL0qWV//W7baD4uIGiVLWQTJXMXUArgB2ATaoGHf3P9b1Q939oDhjtwC31HWfItJA5s6FUaPg7rtDi9Q+fWDQINhll9UvJ7MLafySqRYVAp8A2wP/BIqB/6UwJhFJk4rJbGbQsmW4z8oK9zvaF0ywcyjL6ciKcQXcvfwUOvMpVvggWbvtglnYLplJbio8Nw3J1CDaufsEM+sfrRExNZq7ICIZpLAQ8vKgtDQ8r5iw1sVnMYjh9OFhyliPOzmf0VzO12SvnhlVm5nPKjw3HckkiLLo/jsz6wF8C2ybupBEJB3y89ckB4A9eY98hnI8j/MrGzGWgYxlIN+zda32axYmt2liW9OTTIIYYmZtgEuB24BNgb+nNCoRaXAVdYH9eYvBDKEHz7OYNtzAYG6hPwtpX+d9a6Jb05TMVUzPRg+XAIelNhwRqU+FhdC/f83rJ2SZc6i/ymCG8EdeZQHtuJqhjKMfP9NmnWJQvaHpSuYqponEX3L0rJREJCL1orAQzjwzXGiUmHMULzDYh3Agb/EtWzOQMfyL8yhlo3WOQfWGpi2ZU0zPxjzegNBY79vUhCMi9SU/P3FyMFZxHE+Qz1D24n1KyOYCxjORM1m+5mr2sG3MlUlVJ7hV3MfbRhPdmr5kTjE9FvvczB4GXk5ZRCJSL+LNNWjBSnoziasZxi58zGd04kzu4UFOYSXrrbW9WfhlL81T7bpmBZ0AnVUUaeRiz/2vxwrO5m4+YScKOYVVZHEyD9OF2dzLmXGTQ9V9SPOTTA3iF0INwqL774ErUxyXiMRItthc1QYs5WwmcAWjyOZritibY3mCp+lZY1dV1Q8kmVNMmzREICISX3LF5so24lfO504u40a24gde5w/kUcBL/Jnwt15lVWsLqh8I1JAgzGxDoC+wczRUBDzq7nHXahCR+lddsbmqNizmYm5jADfTjkVM4XB6M4lpHExOjuHFKQ1VMkzCY0wz2w2YDRxE6L9UAvwZeMPMNjOzIQ0SoUgzl0xju/b8yBDyKSGHG7iGN/gD+/E2RzCFaRwCmBrkSa1VdwRxK3Cuu0+JHTSzwwnLg85KZWAizVmyNYet+ZbLuJHz+BcbspRHOYFhXM0HdF1rWxWcpbaqSxBbV00OAO7+spmVEeZDiEg9S6bmkEMxVzCKs5lAC8oppC8juIpP6BJ3+1atVHCW2qsuQWSZ2fruXmnpDzPbAChz99IE7xORdVBdzaETnzGI4ZzCgzjGRM5kJFcyhx0S7q9du7DgmwrOUlvVXed2P2Hlt9yKgejxZOCBVAYl0pzFqxXsykc8RB9m04WT+Td3cCGtvv6K8/xffOU74E7C24IFSg5SNwkThLsPAV4EppnZAjNbAEwFprj7DQ0VoEhzE1sr2JsiHuc4PmJ3/sKzjOZycilmbM4tsK267ktqVTtTxt1vd/dswmpy27t7jrvf1jChiWSO2JXaKlZoS3QrKYE/8DovcCRF7MOhvMZ1XEsOJQxiBL+23lL1BGkQyTTrw91/SXUgIpmq6kptiVdfc7rzCoMZwqFMZT4duJIR3MEF/Gqb4q4JbNKwkkoQIlJ3VVdqW5vzF55lMEPYj3f5ht/Rn5u5i3NZSmsAcrKhuLghohVZoy7N+kSkFhJNUMuinBN4hPfZk2foyRbM5zzuZAe+4lb6r04O1e1DJJVqTBBm1trM/mFmd0XPO5nZX1IfmkjTU1gI7dtXrilUPaXUkjJO5X5msQuPcBIbsIzTuI/f8xkFnMcK1l9rv5rkJumQzBHERGA5cED0fB6gNhsiVVRMcEs0+7kVyzmXAj6lM/dzOstZn5OYxC7M4gFOS9hyW11VJV2SSRAd3X0UUAbg7kuJ1w5SpJlLNMFtQ0q5mFv5ko4UcB4/0oG/8jRdmcEjnMQqWqz1nqzoX2ZODhQUqCgt6ZFMkXpF1NXVAcysI+GIQkRiVK0TbMLPXMAdXMoYtuBHpnIwZzKRlzkcM8O1Ups0cskcQVxLmDC3nZkVAq8AV6Q0KpFGIF49obpbRa1hcxZxDf+kmFxGchXvsRcHMY1DmcrL/Akw1RSkSUhmwaApZvYesD/h1FJ/d1+Q8shE0qgui/R0YD4DGUs/xrEJv/IkxzCUfIrYp9J2apwnTUXCBGFme1UZ+i66zzazbHd/L3VhiaRXbRbp2YZ5XMaN5FHABixjEr0ZxtXMZLe1tlXjPGlKqjuCGFPNaw78sZ5jEWk0kpl3sD1fcSUjOYN7aUE5D3AqI7iKz+i8ehuzsJSnSFOUMEG4+2ENGYhIY5KdHXoixdOZTxjEcPpSSDktuIezGMmVlJAbdz8iTVUyE+U2MLOBZva4mT1mZgOiNSFEMkpsQ714yWF3PuDf9OZjduZEHuFWLmEHvuJC7oibHDR/QZq6ZC5zvR/4Bajo4tqHsB7EiakKSqShVW2oF2tf3iGfofTkGX5mE0ZwFTfxdxbQYa1ts7LCKSU11ZNMkEyC6Ozue8Q8f9XMPkhVQCLpEK+h3kFMYzBDOIIpLKQtY9pcz6VzLuLqzTfn6vSEKdKgkpkH8b6Z7V/xxMz2A95IXUgiDW9NUdo5gpeYxkFM4xD24AMuZxS5FHP5z/+AzTdPZ5giDSqZBLEf8KaZFZtZMfAWcIiZfWRmH9blQ82sv5nNNLNZZjYgGutqZm+b2QwzKzKzfeuyb2nearMwT+wNX0VPnuJd9uUljiSXYi7iNnIp5kYu51c2UcFZmp1kTjEdWZ8faGa7AucC+wIrgBfN7DlgFPBPd3/BzI6Onh9an58tmS35hXnWyKKcE3mEfIayGzP5kh04h7u4n9Moo9Xq7TS5TZqjZGZSl5jZ5sB2sduvw0S5LsDb7l4KYGZTgeMIcys2jbZpA3xbx/1LM1XzwjxrtKSMvhQyiOF05jM+pgun8AD/5mTKq/yz0OQ2aa5qTBBmdgNwBvAlUcM+1m2i3ExgqJm1A5YCRwNFwADgJTO7kXDq68AE8eQBeQDZOuaXGMlMblufZZzJRK5kJLmU8D5dOZ5HeYLj8CpnXDXJTZo78xqOw83sU2A3d19Rbx9qdjbQD/gV+JiQKFoAU939MTM7Cchz98Or20+3bt28qKiovsKSJi43N/Hkttb8Rh4FXM5ofsd3vMX+DGEwz3M0ibrX5+RomU/JTGY23d271bRdMkXqmcBm6x7SGu4+wd33cveDgUXA58DpwOPRJo8QahQia0nUZTVectiUJQxiGMXkchMD+YSd+COvcCBv8jw9SJQcVHMQSa5IPZxwqetMYtaBcPeedf1QM9vC3eebWTbQi7Ba3cXAIcBrhNNXn9d1/5K5ku2y2paF9OcWLuFWNmMJz3MUQ8nnTf5Q42eo5iASJJMg7gNGAh8B9XVG9rGoBlEG9HP3n8zsXOAWM2sJLCOqM4jEqqnL6pZ8z0DGciHj2ZjfeKF1L46adjVH7703RzdcmCIZIZkEscDdb63PD3X3g+KMvQ7sXZ+fI5knUSF6W77mCkZxDnfTihX8m5MZziA+Xrorq/R/lUidJFODmG5mw83sADPbq+KW8sikWalaV2jRIv5kt6rXVOzAlxRwLl/SkfO5k0L60plPOYVCZrGrJreJrINkjiD2jO73jxnTehBSb+LVFSouL010kV0XPuZqhtGHhyljPQrIYzSXM5ec1duom6rIuklmopzWhZCUqs3qbV15n3yG0ovHKaU1YxnIWAbyPVsD6qYqUp+SOYLAzHoAuwCr14Fw9+tTFZQ0L8lMcNuftxjMEHrwPEvYlKHkcwv9WWTtNZlNJEWSWTDoTqA34TJUI6wDkVPtm0QSiG2m17Jl/LrCGs6hvMrLdOctDmQ/3uFqhpLNXK7hBhbSXjUGkRRKpkh9oLufBvzk7v8kzFnYLrVhSSaqaKZXMaGtvDzRls5RPM8b/IFX+SM78zEDGUMuxQznan6mDaDJbCKplkyCWBrdl5rZ7whzF7ZPXUiSqWpqpmes4jgeZzp78zw92IZv6Mc4tmcONzGQ39h49bbt2sE996jGIJJKydQgnjWzzYDRwHuEK5juSmlUkpES1RpasJLeTOJqhrELH/M5O8I995DTty/jWrViXMOGKSKRGo8g3P0Gd1/s7o8Rag87ufs1qQ9NMk3VesF6rOAsJvAJO1HIKThGHx7iyOzZ4brXVq3i70hEGkTCBGFm+5jZVjHPTwMmAzeYWduGCE6apniF6JYt19QeNmApFzKOL9iRCZzDYjbjWJ5gdz7k6dZ9uH5YUhfXiUiKVXcE8S/Cim+Y2cHACOB+YAlQkPrQpClKVIguL4eN+JVLuZE5bM84LmIu2RzJC+yf9T+e4liyc7IoKFBdQaSxqO5PtRbuvih63BsoiE4zPWZmM1IfmjRF8QrRbVjMRdzO37mJdixiCofTm0lM42BycoyVxWkJVURqUG2CMLOW7r4S6E7l7qo6ByBxxRai2/MjA7iZi7idNvzMM/yFoeTzTkzXlmQmyYlIelR3iulhYKqZPUW41PW/AGa2I+E0kwhQuebgDlvxHTdyKcXkMojh/Icj2JP36MkzlZIDrF24FpHGI+GRgLsPNbNXgK2B//iatUmzCLOqRVbXHEpLIZsSrmAUZzOBlqzkIf7GcAbxCV3ivlfN9EQat2pPFbn723HGPktdONLU5OfD70o/ZxDDOZUHcIx7OYORXMlXdFy9XYsWoVBdca9meiKNn2oJUnczZzKsZBi9mcQKWjGeC7mRy5gX04nFDDXTE2mikmm1IVJZUREcdxzsths97Rlu5DJyKWYAt1RKDqAag0hTpgQha4k30a19e+jR5nVesKNgn3346cnXuN6uJdtLGGQjmc+Wa+1HNQaRpk2nmKSS2KIzQHm5051XGLxwCIcylR9pz1UMZzwX8otvGjbyNVcwqcYgkjmUIKSSNRPdnB48x2CGsD/v8A2/YwA3cRfnUspGa73PPSSF4uKGjlhEUkUJQiqZV1LOCTxOPkPpygfMIZfzuJN7OYMVrF/tezXpTSSzKEEIAA/dv5L/DXyYjxhGFz7hEzpzOvfyEH9jJesltQ8VpEUyixJEc7d8Oe9ceB/7TxzJ3/wrPmQ3TmISj3E8q2iR9G5UkBbJPLqKqbkqLYVbb4WOHdnvnvNY4O3oyVN0ZQaPcNLq5NAiyhHt2oVb7FjFfU4O6sIqkoF0BNHc/PILjB8PY8fC/Plw8MEc8c1EpnA4YJU2NYOVK9MTpoikn44gmoHCQth9u5+4zv7JT21y4KqreGl+V/666VTaz5rKFP5E1eQAqimINHc6gshwj46fz3f9b+L1lePYlF940o9hKPkUsQ/8nPh9qimIiBJEpvrmGxg9mh63FrC+L2MyJzGMq/mI3Wt8qya5iQgoQWSeOXNg5EiYOBHKy5nkpzCcQXxG56TebqbJbiISqAaRKT75BE4/HTp1CsnhrLPg88+5LufepJMDqO4gImsoQTRRFQ319rAPeXaj3qzqsjNLH3iEm8svZr8OX3Gh3UHuYdtTUhKOCpKhuoOIxNIppiaosBD+dfa73LJ8KMfwND+XbsIIruJmH8CPbAHfwLt3rNneY5rpVcxlWLQI2rZd8zg7W3UHEalMCaKpmTaN7c4ZwrTlU1jE5lzDP7mNi1nM5tW+Tc30RKS2lCCaAneYMgWGDIH//pfObMEVjOQOLuBXNkl6N2qmJyK1kZYEYWb9gXMJs7PucvebzWwSrK6mbgYsdveu6Yiv0Vi1Cp55JiSGoiK+bbEtI7iVe7LO4bdVG9Z6dypAi0htNHiCMLNdCclhX2AF8KKZPefuvWO2GQMsaejYGo3ycnj00VAU+OgjftliB65qVcDdK04LLbfrsMazCtAiUlvpuIqpC/C2u5e6+0pgKnBcxYtmZsBJwMNpiC29ysrg3nth553h5JPD8wceoOsGnzJ+xblrrcfQokUoPufkwAUXhPtEz9VMT0RqKx2nmGYCQ82sHbAUOBooinn9IOAHd/883pvNLA/IA8jOlHMmy5aFxDBiBJSUQNeu8Mgj0KsXZGUx57T4b1u1KtxERFKhwY8g3H02MBKYArwIfADE9gztQzVHD+5e4O7d3L1bhw4dUhpryv32G9x8M3TsGP7k32orePZZeO89OOEEyAr/eRLlwUzJjyLSOKXhuy58AAAKsElEQVRlopy7T3D3vdz9YGAR8DmAmbUEegGT0hFXg/n5Zxg+PMx0+/vf4fe/h5dfhrfegh49KHzIyM0N+SE3F44+OtQQYqmmICKplpYEYWZbRPfZhIRQccRwOPCJu89LR1wpt3AhXHttKApcfTXssw+8/jq8+ip07w5mFBZCXl440+Qe7u+7L3TRUE1BRBpSuuZBPBbVIMqAfu7+UzR+MplYnP7++7BAz/jx4bTSccdBfj7svfdam+bnh8XeYpWWwvPPa5KbiDSstCQIdz8owfgZDRxKan39NYweDXfdBStWhCuTBg2CXXdN+JZEk9k0yU1EGpqa9aXCl1/CueeG4vMdd8Df/ha6rRYWVkoOFQ33KmoNhYUqSItI46EEUZ8+/hhOPTUUnR94IBQTvvgCJkwIbbhjxKs15OWpIC0ijYcSRH2YMQNOPDEcHTz+eLgyac4cuP32UFGOo7paQ0GBCtIikn7m7umOoc66devmRUVFNW+YKm+9Ff60f+452HRTuPhiGDAA2rev8a1ZWeHIoSozTX4TkdQys+nu3q2m7XQEUVvu8NprcPjhcOCBLJv6NqM3G8LmP5eQ++AQCl+qnBzi1RlAtQYRafzU7jtZ7vDii6Gz6ptvwlZbMf1vN3LUE+fx468bA7A4qiNAOCVUUWeoOJVUEvP60KGVXwPVGkSkcdERRE1WrYInnoBu3UIFed48GDcO5szh+Dcu5celG1favLQ01BcgcZ0hPz8kENUaRKQxUw0ikZUrYfJkGDYMZs2CHXcMcxhOOQVatQJqriOoziAijZFqEHW1YgXccw906bLmz/nCQpg9G846a3VygJrrCKoziEhTpgRRYdmycOqoUyc4++xwVdLjj8OHH4aJbi3XLtcMHVr9nIWaXhcRacyUIH79FcaMge23h4sugm23DZMRiopCz6SsxF9RTXUE1RlEpClrvjWIxYvDEcNNN4Uuq927w+DBcMgh4be5iEiGSrYG0fwuc12wICzSc9ttYV2GHj3CZUUHHJDuyEREGpXmkyC++y6cSrrjDli6FI4/PqzJsOee6Y5MRKRRyvwEUVICo0aFhnllZaHgPGgQ7LxzuiMTEWnUMjdBfP45jBgB998fagpnnAFXXhlacIuISI0yL0F88w1cfjlMmhTmLFxwQXi+3XbpjkxEpEnJvASx4Yahmd5ll8HAgbDllumOSESkScq8eRBt24bFm0eOrHNySNSBVUSkOcm8Iwio1A6jtqrrwKoJbiLSnGTeEcQ6qq4Dq4hIc6IEUcXcubUbFxHJVBmXINa1fqAOrCIiQUYliIr6QUlJWIehon5QmyShDqwiIkFGJYj6qB+oA6uISJBR3Vy1gpuISM2a5Ypyqh+IiNSfjEoQqh+IiNSfjEoQqh+IiNSfjJtJ3bevEoKISH3IqCMIERGpP0oQIiISlxKEiIjEpQQhIiJxKUGIiEhcTXomtZn9CJSkO45Ie2BBuoNoBPQ9rKHvItD3EDSm7yHH3TvUtFGTThCNiZkVJTN1PdPpe1hD30Wg7yFoit+DTjGJiEhcShAiIhKXEkT9KUh3AI2Evoc19F0E+h6CJvc9qAYhIiJx6QhCRETiUoKoR2Y22sw+MbMPzewJM9ss3TGlg5mdaGazzGyVmTWpqzbqg5kdaWafmtkXZnZVuuNJFzO7x8zmm9nMdMeSTma2nZm9amazo38X/dMdU7KUIOrXFGBXd98d+AwYlOZ40mUm0AuYlu5AGpqZtQDGAUcBOwN9zGzn9EaVNvcCR6Y7iEZgJXCpu3cB9gf6NZX/J5Qg6pG7/8fdV0ZP3wa2TWc86eLus93903THkSb7Al+4+1fuvgL4N3BMmmNKC3efBixKdxzp5u7fuft70eNfgNnANumNKjlKEKlzFvBCuoOQBrcN8HXM83k0kV8GknpmlgvsCbyT3kiSk3ELBqWamb0MbBXnpXx3fyraJp9wWFnYkLE1pGS+h2bK4ozpUkHBzDYGHgMGuPvP6Y4nGUoQteTuh1f3upmdDvwF6O4ZfA1xTd9DMzYP2C7m+bbAt2mKRRoJM1uPkBwK3f3xdMeTLJ1iqkdmdiRwJdDT3UvTHY+kxf+ATma2vZm1Ak4Gnk5zTJJGZmbABGC2u49Ndzy1oQRRv24HNgGmmNkMM7sz3QGlg5kdZ2bzgAOA58zspXTH1FCiixQuAl4iFCMnu/us9EaVHmb2MPAW0NnM5pnZ2emOKU3+AJwK/DH6vTDDzI5Od1DJ0ExqERGJS0cQIiISlxKEiIjEpQQhIiJxKUGIiEhcShAiIhKXEoSknZm1i7n873sz+yZ6vNjMPm7gWLrGXoJoZj3r2pHVzIrNrH2c8TZmdr+ZfRndCs1s83WJO8HnJ/xZzOw6M7usvj9TMosShKSduy90967u3hW4E7gpetwVWFXfn2dm1XUQ6Aqs/qXq7k+7+4h6DmEC8JW7d3T3jsAXhM6n9a0hfhbJYEoQ0ti1MLO7oj76/zGzDQHMrKOZvWhm083sv2a2UzSeY2avRGtyvGJm2dH4vWY21sxeBUaa2UbRegX/M7P3zeyYaObz9UDv6Aimt5mdYWa3R/vYMlrn44PodmA0/mQUxywzy6vuhzGzHYG9gRtihq8H9jCzzmZ2qJk9G7P97WZ2RvT4mijemWZWEM3QxcxeM7ORZvaumX1mZgfV9LNUiSnRd3li9FkfmFmza90uShDS+HUCxrn7LsBi4PhovAC42N33Bi4DxkfjtwP3R2tyFAK3xuzr98Dh7n4pkA/8n7vvAxwGjAbWA64BJkVHNJOqxHIrMNXd9wD2AipmSJ8VxdENuMTM2lXz8+wMzHD38oqB6PH7QJcavovb3X0fd98V2JDQ86tCS3ffFxgAXBu1Gq/uZ4mV6Lu8Bvhz9PP2rCE2yUBq1ieN3Rx3nxE9ng7kRl0xDwQeif6IBlg/uj+AsFgRwAPAqJh9PRLzi/kIoGfMefgNgOwaYvkjcBqs/qW+JBq/xMyOix5vR0hqCxPsw4jf3TVeF9iqDjOzK4DWQFtCgnomeq2iAdx0IDeJfYUPrf67fAO418wmx+xfmhElCGnslsc8Lif85ZwFLI7qFDWJ/WX8W8xjA46vurCRme1Xm+DM7FDgcOAAdy81s9cIySaRWcCeZpbl7quifWQBuwPvEZJU7JH9BtE2GxD+su/m7l+b2XVVPqfieyqndv+uE36X7n5+9H30AGaYWVd3T5T4JAPpFJM0OVEv/TlmdiKEbplmtkf08puEDqoAfYHXE+zmJeDimPP4e0bjvxAaLsbzCnBBtH0LM9sUaAP8FCWHnQhLSlYX+xeE00mDY4YHA6+4+1ygBNjZzNY3szZA92ibimSwIPqr/4TqPieJn6UinoTfpZl1dPd33P0aYAGV25hLM6AEIU1VX+BsM/uA8Fd5xbKelwBnmtmHhA6aiRaIv4FQc/jQzGaypmj8KuEX9Awz613lPf0Jp3k+IpzK2QV4EWgZfd4NhKVma3IWoSX4F2b2IyGpnA/g7l8Dk4EPCTWU96PxxcBdwEfAk4S24jWp7meJlei7HG1mH0XfzzTggyQ+UzKIurmKpJGZdQaeJxSJn093PCKxlCBERCQunWISEZG4lCBERCQuJQgREYlLCUJEROJSghARkbiUIEREJC4lCBERiev/A/6+I7EJ7ZxOAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2246ae88d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Quantile-Quantile Plot\n",
    "from statsmodels.graphics.gofplots import qqplot\n",
    "from matplotlib import pyplot\n",
    "\n",
    "# q-q plot\n",
    "qqplot(df.temperature, line='s')\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2246aeac4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# CDF \n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def cdf(data):\n",
    "    # Number of data points: n\n",
    "    n = len(data)\n",
    "\n",
    "    # x-data for the CDF: x\n",
    "    x = np.sort(data)\n",
    "\n",
    "    # y-data for the CDF: y\n",
    "    y = np.arange(1, n+1) / n\n",
    "\n",
    "    return x, y\n",
    "\n",
    "# Sample mean/std\n",
    "mean = np.mean(df.temperature)\n",
    "std  = np.std(df.temperature)\n",
    "# Sampleing the normal distribution with above mean and std\n",
    "normal = np.random.normal(mean, std, 10000)\n",
    "# CDF of the sample and theoretical distributions\n",
    "x, y = cdf(df.temperature)\n",
    "x_norm, y_norm = cdf(normal)\n",
    "# Plot\n",
    "plt.plot(x_norm, y_norm)\n",
    "plt.plot(x, y, marker='*', ls='none')\n",
    "plt.xlabel('Tempature')\n",
    "plt.ylabel('CDF')\n",
    "plt.show()\n",
    "# From graph below for CDF along with P- test and q-q test, its safe to say temperature data is normally distributed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Question 2\n",
    "# Asnwer: Yes the sample size is large enough because the usual threshold for sample size is 30 and we have 130 observation. \n",
    "# Also for sample being independent, it is safe to assume it was different people."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "98.24923076923075"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_temp_means = np.empty(1000)\n",
    "\n",
    "for i in range(1000):\n",
    "    sample_temp=np.random.choice(df.temperature,size=130)\n",
    "    df_temp_means[i]=np.mean(df.temperature)\n",
    "    \n",
    "np.mean(df_temp_means)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We would use one-sample test because the sample set is to be tested against population mean and there is one population for the mean . Z-Test would have been appropraite but we dont have the variance for the population. So, t-test will be more appropriate here. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_1sampResult(statistic=-5.4548232923645195, pvalue=2.4106320415561276e-07)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Question 3\n",
    "# Is the true population mean really 98.6 degrees F?\n",
    "# First, try a bootstrap hypothesis test.\n",
    "# Now, let's try frequentist statistical testing. Would you use a one-sample or two-sample test? Why?\n",
    "# In this situation, is it appropriate to use the  tt  or  zz  statistic?\n",
    "# Now try using the other test. How is the result be different? Why?\n",
    "# T-test\n",
    "stats.ttest_1samp(df.temperature, 98.6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t-score: -5.454823292364301\n",
      "p-value: 2.4106320415585427e-07\n"
     ]
    }
   ],
   "source": [
    "# Since the p-value is really small, T-test proves that our mean temperature is not 98.6 F\n",
    "# Now lets try the t-test manually and compare with scipy.stats function\n",
    "ordered_temps = df.temperature.sort_values()\n",
    "import scipy.stats \n",
    "sample_mean = ordered_temps.mean()\n",
    "pop_mean = 98.6\n",
    "sample_std = ordered_temps.std()\n",
    "sample_size = len(ordered_temps)\n",
    "\n",
    "# calculate t-score\n",
    "t_score = (sample_mean - pop_mean)/(sample_std/np.sqrt(sample_size))\n",
    "\n",
    "# calculate p-value\n",
    "pvalue = scipy.stats.t.sf(np.abs(t_score), sample_size-1)*2\n",
    "\n",
    "print(\"t-score: \" + str(t_score) + \"\\np-value: \" + str(pvalue))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "z-score: -0.4784196512470882\n",
      "p-value: 0.6323515496610721\n"
     ]
    }
   ],
   "source": [
    "# Lets try z-test now\n",
    "import scipy.stats \n",
    "z_score = (sample_mean - pop_mean)/(sample_std)\n",
    "\n",
    "# calculate p-value\n",
    "pvalue_z = scipy.stats.norm.sf(abs(z_score))*2\n",
    "\n",
    "print(\"z-score: \" + str(z_score) + \"\\np-value: \" + str(pvalue_z))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The z-statistics is showing that we should not reject our null hypothesis which is that sample mean is equal to population mean. It gives a different answer from T-statistic. However, T-test is more approriate here and the result of t-test are chosen. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Question 4\n",
    "# Draw a small sample of size 10 from the data and repeat both frequentist tests.\n",
    "# Which one is the correct one to use?\n",
    "# What do you notice? What does this tell you about the difference in application of the  tt  and  zz  statistic?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "z-score: 0.12380157641919584\n",
      "p_value_z: 0.9014723836163081\n",
      "t-score: Ttest_1sampResult(statistic=-1.6179926366078274, pvalue=0.1401181227441729)\n"
     ]
    }
   ],
   "source": [
    "# take a random sample of 10 data points\n",
    "sample = np.random.choice(ordered_temps, size=10)\n",
    "\n",
    "# assign values needed for formulas\n",
    "mu = ordered_temps.mean()\n",
    "std = ordered_temps.std()\n",
    "mean_10 = sample.mean()\n",
    "std_10 = sample.std()\n",
    "\n",
    "# calculate z- and t-scores\n",
    "z_10 = (mean_10 - mu)/(std)\n",
    "p_value_z10=scipy.stats.norm.sf(abs(z_10))*2\n",
    "#t_10 = (mean_10 - mu)/(std/np.sqrt(10))\n",
    "t_10=scipy.stats.ttest_1samp(sample,98.6)\n",
    "\n",
    "print(\"z-score: \" + str(z_10) + \"\\np_value_z: \" + str(p_value_z10) + \"\\nt-score: \" + str(t_10) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "95% confidence Interval for Bootstrap: (96.81775582674824, 99.68070571171332)\n",
      "95% confidence interval for Frequentist: (96.81775582674824, 99.68070571171332) \n"
     ]
    }
   ],
   "source": [
    "# Question 5\n",
    "# At what temperature should we consider someone's temperature to be \"abnormal\"?\n",
    "# As in the previous example, try calculating everything using the boostrap approach, as well as the frequentist approach.\n",
    "# Start by computing the margin of error and confidence interval. \n",
    "# When calculating the confidence interval, keep in mind that you should use the appropriate formula for one draw, \n",
    "# and not N draws.\n",
    "\n",
    "# sample mean and standard deviation for use with bootstrap and frequentist approaches below\n",
    "x_bar = np.mean(df.temperature)\n",
    "s = np.std(df.temperature)\n",
    "\n",
    "# Bootstrap Approach \n",
    "temp_means = np.empty(1000)\n",
    "\n",
    "for i in range(1000):\n",
    "    samp_temp=np.random.choice(df.temperature,size=130)\n",
    "    temp_means[i]=np.mean(df.temperature)\n",
    "s=np.std(df.temperature)    \n",
    "ci = scipy.stats.norm.interval(0.95, loc=x_bar, scale=s)\n",
    "print(\"95% confidence Interval for Bootstrap: \" + str(ci))\n",
    "\n",
    "# frequentist approach - confidence interval for the one-sample t-test\n",
    "\n",
    "# alpha = 0.05, confidence coefficient = 95%\n",
    "\n",
    "# confidence interval for one draw\n",
    "ci_low_f, ci_high_f = stats.norm.interval(0.95, loc=x_bar, scale=s)\n",
    "print('95% confidence interval for Frequentist: ({:}, {:}) '.format(ci_low_f, ci_high_f))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Both approaches gave the same 95% confidence interval and above or below these temperatures would be considered abnormal. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Of the 130 participants, 65 are female and 65 are male.\n"
     ]
    }
   ],
   "source": [
    "# Question 6 \n",
    "# Is there a significant difference between males and females in normal temperature?\n",
    "# What testing approach did you use and why?\n",
    "# Write a story with your conclusion in the context of the original problem.\n",
    "\n",
    "\n",
    "males = df[df.gender == 'M']\n",
    "females = df[df.gender == 'F']\n",
    "\n",
    "print('Of the {} participants, {} are female and {} are male.'.format(len(males + females), len(females), len(males)))\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2246af7c860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# boxplot for visualizing temperature w.r.t. males and females\n",
    "import seaborn as sns\n",
    "sns.boxplot(x = 'gender', y = 'temperature', data=df)\n",
    "sns.set(rc={\"figure.figsize\": (12, 8)})\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "plt.xlabel('Gender')\n",
    "plt.ylabel('Temp')\n",
    "plt.title('Fig. 6.1: Body Temp by Gender')\n",
    "\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2246ae88e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Checking if samples normally distributed?\n",
    "sns.set(rc={\"figure.figsize\": (12, 8)})\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "# Calculating the CDFs for males and females\n",
    "x_male, y_male = cdf(males.temperature)\n",
    "x_female, y_female = cdf(females.temperature)\n",
    "\n",
    "# Generate plot\n",
    "plt.plot(x_male, y_male, marker = '.', linestyle = 'none', color='b')\n",
    "plt.plot(x_female, y_female, marker='.', linestyle='none', color='r')\n",
    "\n",
    "# draw 100,000 random samples from a normal distribution\n",
    "m_norm_dist = np.random.normal(np.mean(males.temperature), np.std(males.temperature), 100000)\n",
    "mnd_x, mnd_y = cdf(m_norm_dist)\n",
    "plt.plot(mnd_x, mnd_y, color='b', alpha=0.5)\n",
    "\n",
    "# draw 100,000 random samples from a normal distribution\n",
    "f_norm_dist = np.random.normal(np.mean(females.temperature), np.std(females.temperature), 100000)\n",
    "fnd_x, fnd_y = cdf(f_norm_dist)\n",
    "plt.plot(fnd_x, fnd_y, color='r', alpha=0.5)\n",
    "\n",
    "# Make the margins nice\n",
    "plt.margins = 0.02\n",
    "\n",
    "# Label the axes\n",
    "plt.xlabel('Gender')\n",
    "plt.ylabel('CDF')\n",
    "plt.title('Fig. 6.2: CDFs of males v. females')\n",
    "plt.legend(('males', 'females'), loc='lower right', fontsize='large', markerscale=2)\n",
    "\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# common variables\n",
    "temp_m = males.temperature\n",
    "temp_f = females.temperature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Male variance: 0.4807479289940825\n",
      "Female variance: 0.5442698224852062\n"
     ]
    }
   ],
   "source": [
    "# Check for identical variances\n",
    "mv = np.var(temp_m)\n",
    "fv = np.var(temp_f)\n",
    "\n",
    "print('Male variance: {}\\nFemale variance: {}'.format(mv, fv))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bootstrap verification:\n",
      "Male variance: 0.4744482404733727\n",
      "Female variance: 0.5353292970414192\n"
     ]
    }
   ],
   "source": [
    "# confirm that variances are not equal with bootstrap - null hypothesis is that they are equal\n",
    "\n",
    "size = 10000\n",
    "\n",
    "bs_replicates_m = np.empty(size)\n",
    "\n",
    "for i in range(size):\n",
    "    bs_sample_m = np.random.choice(temp_m, len(temp_m))\n",
    "    bs_replicates_m[i] = np.var(bs_sample_m)\n",
    "    \n",
    "bs_var_m = np.sum(bs_replicates_m)/size\n",
    "\n",
    "bs_replicates_f = np.empty(size)\n",
    "\n",
    "for i in range(size):\n",
    "    bs_sample_f = np.random.choice(temp_f, len(temp_f))\n",
    "    bs_replicates_f[i] = np.var(bs_sample_f)\n",
    "\n",
    "bs_var_f = np.sum(bs_replicates_f)/size\n",
    "\n",
    "bs_var_m\n",
    "bs_var_f\n",
    "print('Bootstrap verification:\\nMale variance: {}\\nFemale variance: {}'.format(bs_var_m, bs_var_f))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "t-statistic: -2.285\n",
      "p-value: 0.02394\n"
     ]
    }
   ],
   "source": [
    "# Variances are not identical, so set `equal_var` to false to perform Welch's t-test\n",
    "r = stats.ttest_ind(temp_m, temp_f, equal_var=False)\n",
    "print('t-statistic: {:0.4}\\np-value: {:0.4}'.format(r.statistic, r.pvalue))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MALES\n",
      "sample mean: 98.105\n",
      "sample standard deviation: 0.693\n",
      "\n",
      "95% confidence interval (one draw): 96.746 - 99.464\n",
      "\n",
      "\n",
      "FEMALES\n",
      "sample mean: 98.394\n",
      "sample standard deviation: 0.738\n",
      "\n",
      "95% confidence interval (one draw): 96.948 - 99.84\n"
     ]
    }
   ],
   "source": [
    "# males\n",
    "print('MALES')\n",
    "xbar_m = np.mean(temp_m)\n",
    "s_m = np.std(temp_m)\n",
    "print('sample mean: {}\\nsample standard deviation: {}\\n'.format(round(xbar_m, 3), round(s_m, 3)))\n",
    "\n",
    "# confidence interval for one draw\n",
    "ci_low_m, ci_high_m = stats.norm.interval(0.95, loc=xbar_m, scale=s_m)\n",
    "print('95% confidence interval (one draw): {} - {}'.format(round(ci_low_m, 3), round(ci_high_m, 3)))\n",
    "\n",
    "# females\n",
    "print('\\n\\nFEMALES')\n",
    "xbar_f = np.mean(temp_f)\n",
    "s_f = np.std(temp_f)\n",
    "print('sample mean: {}\\nsample standard deviation: {}\\n'.format(round(xbar_f, 3), round(s_f, 3)))\n",
    "\n",
    "# confidence interval for one draw\n",
    "ci_low_f, ci_high_f = stats.norm.interval(0.95, loc=xbar_f, scale=s_f)\n",
    "print('95% confidence interval (one draw): {} - {}'.format(round(ci_low_f, 3), round(ci_high_f, 3)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Welch's t-test:\n",
      "t-statistic: 227.6\n",
      "p-value: 0.0\n"
     ]
    }
   ],
   "source": [
    "# bootstrap - two-sided Welch's t-test  \n",
    "size = 10000\n",
    "bs_replicates_m = np.empty(size)\n",
    "\n",
    "for i in range(size):\n",
    "    bs_sample_m = np.random.choice(temp_m, len(temp_m))\n",
    "    bs_replicates_m[i] = np.mean(bs_sample_m)\n",
    "    \n",
    "bs_mean_m = np.sum(bs_replicates_m)/size\n",
    "\n",
    "bs_replicates_f = np.empty(size)\n",
    "\n",
    "for i in range(size):\n",
    "    bs_sample_f = np.random.choice(temp_f, len(temp_f))\n",
    "    bs_replicates_f[i] = np.mean(bs_sample_f)\n",
    "\n",
    "bs_mean_f = np.sum(bs_replicates_f)/size\n",
    "\n",
    "\n",
    "result = stats.ttest_ind(bs_replicates_f, bs_replicates_m, equal_var=False)\n",
    "\n",
    "print('Welch\\'s t-test:\\nt-statistic: {:0.5}\\np-value: {:0.5}'.format(result[0], result[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Results: Overall from the boxplot we can see that females have a slightly higher body temperature then men. Frome our tests the mean human body temperature is not 98.6 degF but 98.25 degF. Also both samples follow normal distribution. "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}