{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": [
     "remove_input"
    ]
   },
   "outputs": [],
   "source": [
    "path_data = '../../data/'\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Bootstrap\n",
    "A data scientist is using the data in a random sample to estimate an unknown parameter. She uses the sample to calculate the value of a statistic that she will use as her estimate. \n",
    "\n",
    "Once she has calculated the observed value of her statistic, she could just present it as her estimate and go on her merry way. But she's a data scientist. She knows that her random sample is just one of numerous possible random samples, and thus her estimate is just one of numerous plausible estimates. \n",
    "\n",
    "By how much could those estimates vary? To answer this, it appears as though she needs to draw another sample from the population, and compute a new estimate based on the new sample. But she doesn't have the resources to go back to the population and draw another sample.\n",
    "\n",
    "It looks as though the data scientist is stuck.\n",
    "\n",
    "Fortunately, a brilliant idea called *the bootstrap* can help her out. Since it is not feasible to generate new samples from the population, the bootstrap generates new random samples by a method called *resampling*: the new samples are drawn at random *from the original sample*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, we will see how and why the bootstrap works. In the rest of the chapter, we will use the bootstrap for inference.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Employee Compensation in the City of San Francisco\n",
    "[SF OpenData](https://data.sfgov.org) is a website where the City and County of San Francisco make some of their data publicly available. One of the data sets contains compensation data for employees of the City. These include medical professionals at City-run hospitals, police officers, fire fighters, transportation workers, elected officials, and all other employees of the City. \n",
    "\n",
    "Compensation data for the calendar year 2015 are in the table `sf2015`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sf2015 = pd.read_csv(path_data + 'san_francisco_2015.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year Type</th>\n",
       "      <th>Year</th>\n",
       "      <th>Organization Group Code</th>\n",
       "      <th>Organization Group</th>\n",
       "      <th>Department Code</th>\n",
       "      <th>Department</th>\n",
       "      <th>Union Code</th>\n",
       "      <th>Union</th>\n",
       "      <th>Job Family Code</th>\n",
       "      <th>Job Family</th>\n",
       "      <th>...</th>\n",
       "      <th>Employee Identifier</th>\n",
       "      <th>Salaries</th>\n",
       "      <th>Overtime</th>\n",
       "      <th>Other Salaries</th>\n",
       "      <th>Total Salary</th>\n",
       "      <th>Retirement</th>\n",
       "      <th>Health/Dental</th>\n",
       "      <th>Other Benefits</th>\n",
       "      <th>Total Benefits</th>\n",
       "      <th>Total Compensation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>Public Works, Transportation &amp; Commerce</td>\n",
       "      <td>WTR</td>\n",
       "      <td>PUC Water Department</td>\n",
       "      <td>21.0</td>\n",
       "      <td>Prof &amp; Tech Engineers - Miscellaneous, Local 21</td>\n",
       "      <td>2400</td>\n",
       "      <td>Lab, Pharmacy &amp; Med Techs</td>\n",
       "      <td>...</td>\n",
       "      <td>21538</td>\n",
       "      <td>82146.04</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>82146.04</td>\n",
       "      <td>16942.21</td>\n",
       "      <td>12340.88</td>\n",
       "      <td>6337.73</td>\n",
       "      <td>35620.82</td>\n",
       "      <td>117766.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>Public Works, Transportation &amp; Commerce</td>\n",
       "      <td>DPW</td>\n",
       "      <td>General Services Agency - Public Works</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Carpet, Linoleum and Soft Tile Workers, Local 12</td>\n",
       "      <td>7300</td>\n",
       "      <td>Journeyman Trade</td>\n",
       "      <td>...</td>\n",
       "      <td>5459</td>\n",
       "      <td>32165.75</td>\n",
       "      <td>973.19</td>\n",
       "      <td>848.96</td>\n",
       "      <td>33987.90</td>\n",
       "      <td>0.00</td>\n",
       "      <td>4587.51</td>\n",
       "      <td>2634.42</td>\n",
       "      <td>7221.93</td>\n",
       "      <td>41209.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>790.0</td>\n",
       "      <td>SEIU - Miscellaneous, Local 1021</td>\n",
       "      <td>1600</td>\n",
       "      <td>Payroll, Billing &amp; Accounting</td>\n",
       "      <td>...</td>\n",
       "      <td>41541</td>\n",
       "      <td>71311.00</td>\n",
       "      <td>5757.98</td>\n",
       "      <td>0.00</td>\n",
       "      <td>77068.98</td>\n",
       "      <td>14697.59</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>6370.06</td>\n",
       "      <td>33492.15</td>\n",
       "      <td>110561.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>351.0</td>\n",
       "      <td>Municipal Executive Association - Miscellaneous</td>\n",
       "      <td>0900</td>\n",
       "      <td>Management</td>\n",
       "      <td>...</td>\n",
       "      <td>26718</td>\n",
       "      <td>28430.25</td>\n",
       "      <td>0.00</td>\n",
       "      <td>763.07</td>\n",
       "      <td>29193.32</td>\n",
       "      <td>0.00</td>\n",
       "      <td>4223.14</td>\n",
       "      <td>5208.51</td>\n",
       "      <td>9431.65</td>\n",
       "      <td>38624.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>Public Works, Transportation &amp; Commerce</td>\n",
       "      <td>MTA</td>\n",
       "      <td>Municipal Transportation Agency</td>\n",
       "      <td>790.0</td>\n",
       "      <td>SEIU - Miscellaneous, Local 1021</td>\n",
       "      <td>8200</td>\n",
       "      <td>Protection &amp; Apprehension</td>\n",
       "      <td>...</td>\n",
       "      <td>45810</td>\n",
       "      <td>7948.75</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>7948.75</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2873.17</td>\n",
       "      <td>616.24</td>\n",
       "      <td>3489.41</td>\n",
       "      <td>11438.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>Public Protection</td>\n",
       "      <td>POL</td>\n",
       "      <td>Police</td>\n",
       "      <td>911.0</td>\n",
       "      <td>Police Officers' Association</td>\n",
       "      <td>Q000</td>\n",
       "      <td>Police Services</td>\n",
       "      <td>...</td>\n",
       "      <td>32906</td>\n",
       "      <td>2235.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2235.00</td>\n",
       "      <td>490.36</td>\n",
       "      <td>286.72</td>\n",
       "      <td>176.57</td>\n",
       "      <td>953.65</td>\n",
       "      <td>3188.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>791.0</td>\n",
       "      <td>SEIU - Staff and Per Diem Nurses, Local 1021</td>\n",
       "      <td>2300</td>\n",
       "      <td>Nursing</td>\n",
       "      <td>...</td>\n",
       "      <td>7506</td>\n",
       "      <td>187247.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>11704.06</td>\n",
       "      <td>198951.06</td>\n",
       "      <td>37683.66</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>11221.73</td>\n",
       "      <td>61329.89</td>\n",
       "      <td>260280.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>Public Works, Transportation &amp; Commerce</td>\n",
       "      <td>MTA</td>\n",
       "      <td>Municipal Transportation Agency</td>\n",
       "      <td>253.0</td>\n",
       "      <td>Transport Workers - Transit Operators, Local 2...</td>\n",
       "      <td>9100</td>\n",
       "      <td>Street Transit</td>\n",
       "      <td>...</td>\n",
       "      <td>36773</td>\n",
       "      <td>66988.54</td>\n",
       "      <td>3512.88</td>\n",
       "      <td>2770.39</td>\n",
       "      <td>73271.81</td>\n",
       "      <td>19127.22</td>\n",
       "      <td>13202.95</td>\n",
       "      <td>5455.10</td>\n",
       "      <td>37785.27</td>\n",
       "      <td>111057.08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>CAT</td>\n",
       "      <td>City Attorney</td>\n",
       "      <td>311.0</td>\n",
       "      <td>Municipal Attorneys' Association</td>\n",
       "      <td>8100</td>\n",
       "      <td>Legal &amp; Court</td>\n",
       "      <td>...</td>\n",
       "      <td>12963</td>\n",
       "      <td>135189.60</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1562.50</td>\n",
       "      <td>136752.10</td>\n",
       "      <td>27501.81</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>10102.97</td>\n",
       "      <td>50029.28</td>\n",
       "      <td>186781.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>3</td>\n",
       "      <td>Human Welfare &amp; Neighborhood Development</td>\n",
       "      <td>DSS</td>\n",
       "      <td>Human Services</td>\n",
       "      <td>535.0</td>\n",
       "      <td>SEIU - Human Services, Local 1021</td>\n",
       "      <td>9700</td>\n",
       "      <td>Community Development</td>\n",
       "      <td>...</td>\n",
       "      <td>35179</td>\n",
       "      <td>70474.77</td>\n",
       "      <td>147.28</td>\n",
       "      <td>1647.24</td>\n",
       "      <td>72269.29</td>\n",
       "      <td>14650.28</td>\n",
       "      <td>10696.90</td>\n",
       "      <td>5993.11</td>\n",
       "      <td>31340.29</td>\n",
       "      <td>103609.58</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  Year Type  Year  Organization Group Code  \\\n",
       "0  Calendar  2015                        2   \n",
       "1  Calendar  2015                        2   \n",
       "2  Calendar  2015                        4   \n",
       "3  Calendar  2015                        4   \n",
       "4  Calendar  2015                        2   \n",
       "5  Calendar  2015                        1   \n",
       "6  Calendar  2015                        4   \n",
       "7  Calendar  2015                        2   \n",
       "8  Calendar  2015                        6   \n",
       "9  Calendar  2015                        3   \n",
       "\n",
       "                         Organization Group Department Code  \\\n",
       "0   Public Works, Transportation & Commerce             WTR   \n",
       "1   Public Works, Transportation & Commerce             DPW   \n",
       "2                          Community Health             DPH   \n",
       "3                          Community Health             DPH   \n",
       "4   Public Works, Transportation & Commerce             MTA   \n",
       "5                         Public Protection             POL   \n",
       "6                          Community Health             DPH   \n",
       "7   Public Works, Transportation & Commerce             MTA   \n",
       "8          General Administration & Finance             CAT   \n",
       "9  Human Welfare & Neighborhood Development             DSS   \n",
       "\n",
       "                               Department  Union Code  \\\n",
       "0                    PUC Water Department        21.0   \n",
       "1  General Services Agency - Public Works        12.0   \n",
       "2                           Public Health       790.0   \n",
       "3                           Public Health       351.0   \n",
       "4         Municipal Transportation Agency       790.0   \n",
       "5                                  Police       911.0   \n",
       "6                           Public Health       791.0   \n",
       "7         Municipal Transportation Agency       253.0   \n",
       "8                           City Attorney       311.0   \n",
       "9                          Human Services       535.0   \n",
       "\n",
       "                                               Union Job Family Code  \\\n",
       "0    Prof & Tech Engineers - Miscellaneous, Local 21            2400   \n",
       "1   Carpet, Linoleum and Soft Tile Workers, Local 12            7300   \n",
       "2                   SEIU - Miscellaneous, Local 1021            1600   \n",
       "3    Municipal Executive Association - Miscellaneous            0900   \n",
       "4                   SEIU - Miscellaneous, Local 1021            8200   \n",
       "5                       Police Officers' Association            Q000   \n",
       "6       SEIU - Staff and Per Diem Nurses, Local 1021            2300   \n",
       "7  Transport Workers - Transit Operators, Local 2...            9100   \n",
       "8                   Municipal Attorneys' Association            8100   \n",
       "9                  SEIU - Human Services, Local 1021            9700   \n",
       "\n",
       "                      Job Family  ... Employee Identifier   Salaries  \\\n",
       "0      Lab, Pharmacy & Med Techs  ...               21538   82146.04   \n",
       "1               Journeyman Trade  ...                5459   32165.75   \n",
       "2  Payroll, Billing & Accounting  ...               41541   71311.00   \n",
       "3                     Management  ...               26718   28430.25   \n",
       "4      Protection & Apprehension  ...               45810    7948.75   \n",
       "5                Police Services  ...               32906    2235.00   \n",
       "6                        Nursing  ...                7506  187247.00   \n",
       "7                 Street Transit  ...               36773   66988.54   \n",
       "8                  Legal & Court  ...               12963  135189.60   \n",
       "9          Community Development  ...               35179   70474.77   \n",
       "\n",
       "   Overtime  Other Salaries  Total Salary  Retirement  Health/Dental  \\\n",
       "0      0.00            0.00      82146.04    16942.21       12340.88   \n",
       "1    973.19          848.96      33987.90        0.00        4587.51   \n",
       "2   5757.98            0.00      77068.98    14697.59       12424.50   \n",
       "3      0.00          763.07      29193.32        0.00        4223.14   \n",
       "4      0.00            0.00       7948.75        0.00        2873.17   \n",
       "5      0.00            0.00       2235.00      490.36         286.72   \n",
       "6      0.00        11704.06     198951.06    37683.66       12424.50   \n",
       "7   3512.88         2770.39      73271.81    19127.22       13202.95   \n",
       "8      0.00         1562.50     136752.10    27501.81       12424.50   \n",
       "9    147.28         1647.24      72269.29    14650.28       10696.90   \n",
       "\n",
       "   Other Benefits  Total Benefits  Total Compensation  \n",
       "0         6337.73        35620.82           117766.86  \n",
       "1         2634.42         7221.93            41209.83  \n",
       "2         6370.06        33492.15           110561.13  \n",
       "3         5208.51         9431.65            38624.97  \n",
       "4          616.24         3489.41            11438.16  \n",
       "5          176.57          953.65             3188.65  \n",
       "6        11221.73        61329.89           260280.95  \n",
       "7         5455.10        37785.27           111057.08  \n",
       "8        10102.97        50029.28           186781.38  \n",
       "9         5993.11        31340.29           103609.58  \n",
       "\n",
       "[10 rows x 22 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sf2015.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is one row for each of 42,979 employees. There are numerous columns containing information about City departmental affiliation and details of the different parts of the employee's compensation package. Here is the row correspoding to the late Edward Lee, the Mayor at that time."
   ]
  },
  {
   "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>Year Type</th>\n",
       "      <th>Year</th>\n",
       "      <th>Organization Group Code</th>\n",
       "      <th>Organization Group</th>\n",
       "      <th>Department Code</th>\n",
       "      <th>Department</th>\n",
       "      <th>Union Code</th>\n",
       "      <th>Union</th>\n",
       "      <th>Job Family Code</th>\n",
       "      <th>Job Family</th>\n",
       "      <th>...</th>\n",
       "      <th>Employee Identifier</th>\n",
       "      <th>Salaries</th>\n",
       "      <th>Overtime</th>\n",
       "      <th>Other Salaries</th>\n",
       "      <th>Total Salary</th>\n",
       "      <th>Retirement</th>\n",
       "      <th>Health/Dental</th>\n",
       "      <th>Other Benefits</th>\n",
       "      <th>Total Benefits</th>\n",
       "      <th>Total Compensation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3335</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>MYR</td>\n",
       "      <td>Mayor</td>\n",
       "      <td>556.0</td>\n",
       "      <td>Elected Officials</td>\n",
       "      <td>1100</td>\n",
       "      <td>Administrative &amp; Mgmt (Unrep)</td>\n",
       "      <td>...</td>\n",
       "      <td>22433</td>\n",
       "      <td>288963.55</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>288963.55</td>\n",
       "      <td>58117.03</td>\n",
       "      <td>12424.5</td>\n",
       "      <td>20292.95</td>\n",
       "      <td>90834.48</td>\n",
       "      <td>379798.03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Year Type  Year  Organization Group Code  \\\n",
       "3335  Calendar  2015                        6   \n",
       "\n",
       "                    Organization Group Department Code Department  Union Code  \\\n",
       "3335  General Administration & Finance             MYR      Mayor       556.0   \n",
       "\n",
       "                  Union Job Family Code                     Job Family  ...  \\\n",
       "3335  Elected Officials            1100  Administrative & Mgmt (Unrep)  ...   \n",
       "\n",
       "     Employee Identifier   Salaries  Overtime  Other Salaries  Total Salary  \\\n",
       "3335               22433  288963.55       0.0             0.0     288963.55   \n",
       "\n",
       "      Retirement  Health/Dental  Other Benefits  Total Benefits  \\\n",
       "3335    58117.03        12424.5        20292.95        90834.48   \n",
       "\n",
       "      Total Compensation  \n",
       "3335           379798.03  \n",
       "\n",
       "[1 rows x 22 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sf2015[sf2015['Job'] == 'Mayor']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are going to study the final column, `Total Compensation`. That's the employee's salary plus the City's contribution towards his/her retirement and benefit plans.\n",
    "\n",
    "Financial packages in a calendar year can sometimes be hard to understand as they depend on the date of hire, whether the employee is changing jobs within the City, and so on. For example, the lowest values in the `Total Compensation` column look a little strange."
   ]
  },
  {
   "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>Year Type</th>\n",
       "      <th>Year</th>\n",
       "      <th>Organization Group Code</th>\n",
       "      <th>Organization Group</th>\n",
       "      <th>Department Code</th>\n",
       "      <th>Department</th>\n",
       "      <th>Union Code</th>\n",
       "      <th>Union</th>\n",
       "      <th>Job Family Code</th>\n",
       "      <th>Job Family</th>\n",
       "      <th>...</th>\n",
       "      <th>Employee Identifier</th>\n",
       "      <th>Salaries</th>\n",
       "      <th>Overtime</th>\n",
       "      <th>Other Salaries</th>\n",
       "      <th>Total Salary</th>\n",
       "      <th>Retirement</th>\n",
       "      <th>Health/Dental</th>\n",
       "      <th>Other Benefits</th>\n",
       "      <th>Total Benefits</th>\n",
       "      <th>Total Compensation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27308</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>Public Protection</td>\n",
       "      <td>FIR</td>\n",
       "      <td>Fire Department</td>\n",
       "      <td>798.0</td>\n",
       "      <td>Firefighters - Miscellaneous, Local 798</td>\n",
       "      <td>H000</td>\n",
       "      <td>Fire Services</td>\n",
       "      <td>...</td>\n",
       "      <td>43833</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-423.76</td>\n",
       "      <td>-423.76</td>\n",
       "      <td>-423.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15746</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>790.0</td>\n",
       "      <td>SEIU - Miscellaneous, Local 1021</td>\n",
       "      <td>9900</td>\n",
       "      <td>Public Service Aide</td>\n",
       "      <td>...</td>\n",
       "      <td>27871</td>\n",
       "      <td>-292.40</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-292.40</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-95.58</td>\n",
       "      <td>-22.63</td>\n",
       "      <td>-118.21</td>\n",
       "      <td>-410.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24576</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>Public Protection</td>\n",
       "      <td>JUV</td>\n",
       "      <td>Juvenile Probation</td>\n",
       "      <td>790.0</td>\n",
       "      <td>SEIU - Miscellaneous, Local 1021</td>\n",
       "      <td>8300</td>\n",
       "      <td>Correction &amp; Detention</td>\n",
       "      <td>...</td>\n",
       "      <td>10517</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-159.12</td>\n",
       "      <td>-159.12</td>\n",
       "      <td>-159.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42982</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>CPC</td>\n",
       "      <td>City Planning</td>\n",
       "      <td>21.0</td>\n",
       "      <td>Prof &amp; Tech Engineers - Miscellaneous, Local 21</td>\n",
       "      <td>1000</td>\n",
       "      <td>Information Systems</td>\n",
       "      <td>...</td>\n",
       "      <td>18961</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-26.53</td>\n",
       "      <td>-26.53</td>\n",
       "      <td>-26.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23310</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>CPC</td>\n",
       "      <td>City Planning</td>\n",
       "      <td>21.0</td>\n",
       "      <td>Prof &amp; Tech Engineers - Miscellaneous, Local 21</td>\n",
       "      <td>5200</td>\n",
       "      <td>Professional Engineering</td>\n",
       "      <td>...</td>\n",
       "      <td>19387</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-9.51</td>\n",
       "      <td>-9.51</td>\n",
       "      <td>-9.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5171</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>351.0</td>\n",
       "      <td>Municipal Executive Association - Miscellaneous</td>\n",
       "      <td>0900</td>\n",
       "      <td>Management</td>\n",
       "      <td>...</td>\n",
       "      <td>1523</td>\n",
       "      <td>256098.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>82292.31</td>\n",
       "      <td>338390.32</td>\n",
       "      <td>51977.53</td>\n",
       "      <td>11468.77</td>\n",
       "      <td>20963.32</td>\n",
       "      <td>84409.62</td>\n",
       "      <td>422799.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17805</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>Public Works, Transportation &amp; Commerce</td>\n",
       "      <td>AIR</td>\n",
       "      <td>Airport Commission</td>\n",
       "      <td>351.0</td>\n",
       "      <td>Municipal Executive Association - Miscellaneous</td>\n",
       "      <td>0900</td>\n",
       "      <td>Management</td>\n",
       "      <td>...</td>\n",
       "      <td>17356</td>\n",
       "      <td>326764.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>326764.01</td>\n",
       "      <td>65806.33</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>21691.23</td>\n",
       "      <td>99922.06</td>\n",
       "      <td>426686.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>499</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>ADM</td>\n",
       "      <td>General Services Agency - City Admin</td>\n",
       "      <td>164.0</td>\n",
       "      <td>Physicians and Dentists - Miscellaneous</td>\n",
       "      <td>2500</td>\n",
       "      <td>Med Therapy &amp; Auxiliary</td>\n",
       "      <td>...</td>\n",
       "      <td>13746</td>\n",
       "      <td>279311.03</td>\n",
       "      <td>9046.92</td>\n",
       "      <td>56742.56</td>\n",
       "      <td>345100.51</td>\n",
       "      <td>56211.61</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>13482.66</td>\n",
       "      <td>82118.77</td>\n",
       "      <td>427219.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13194</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>ADM</td>\n",
       "      <td>General Services Agency - City Admin</td>\n",
       "      <td>164.0</td>\n",
       "      <td>Physicians and Dentists - Miscellaneous</td>\n",
       "      <td>2500</td>\n",
       "      <td>Med Therapy &amp; Auxiliary</td>\n",
       "      <td>...</td>\n",
       "      <td>1016</td>\n",
       "      <td>279311.10</td>\n",
       "      <td>3829.36</td>\n",
       "      <td>114433.58</td>\n",
       "      <td>397574.04</td>\n",
       "      <td>56211.64</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>14299.10</td>\n",
       "      <td>82935.24</td>\n",
       "      <td>480509.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19177</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>RET</td>\n",
       "      <td>Retirement System</td>\n",
       "      <td>351.0</td>\n",
       "      <td>Municipal Executive Association - Miscellaneous</td>\n",
       "      <td>1100</td>\n",
       "      <td>Administrative &amp; Mgmt (Unrep)</td>\n",
       "      <td>...</td>\n",
       "      <td>46881</td>\n",
       "      <td>507831.60</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>507831.60</td>\n",
       "      <td>105052.98</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>23566.16</td>\n",
       "      <td>141043.64</td>\n",
       "      <td>648875.24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>42989 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Year Type  Year  Organization Group Code  \\\n",
       "27308  Calendar  2015                        1   \n",
       "15746  Calendar  2015                        4   \n",
       "24576  Calendar  2015                        1   \n",
       "42982  Calendar  2015                        6   \n",
       "23310  Calendar  2015                        6   \n",
       "...         ...   ...                      ...   \n",
       "5171   Calendar  2015                        4   \n",
       "17805  Calendar  2015                        2   \n",
       "499    Calendar  2015                        6   \n",
       "13194  Calendar  2015                        6   \n",
       "19177  Calendar  2015                        6   \n",
       "\n",
       "                            Organization Group Department Code  \\\n",
       "27308                        Public Protection             FIR   \n",
       "15746                         Community Health             DPH   \n",
       "24576                        Public Protection             JUV   \n",
       "42982         General Administration & Finance             CPC   \n",
       "23310         General Administration & Finance             CPC   \n",
       "...                                        ...             ...   \n",
       "5171                          Community Health             DPH   \n",
       "17805  Public Works, Transportation & Commerce             AIR   \n",
       "499           General Administration & Finance             ADM   \n",
       "13194         General Administration & Finance             ADM   \n",
       "19177         General Administration & Finance             RET   \n",
       "\n",
       "                                 Department  Union Code  \\\n",
       "27308                       Fire Department       798.0   \n",
       "15746                         Public Health       790.0   \n",
       "24576                    Juvenile Probation       790.0   \n",
       "42982                         City Planning        21.0   \n",
       "23310                         City Planning        21.0   \n",
       "...                                     ...         ...   \n",
       "5171                          Public Health       351.0   \n",
       "17805                    Airport Commission       351.0   \n",
       "499    General Services Agency - City Admin       164.0   \n",
       "13194  General Services Agency - City Admin       164.0   \n",
       "19177                     Retirement System       351.0   \n",
       "\n",
       "                                                 Union Job Family Code  \\\n",
       "27308          Firefighters - Miscellaneous, Local 798            H000   \n",
       "15746                 SEIU - Miscellaneous, Local 1021            9900   \n",
       "24576                 SEIU - Miscellaneous, Local 1021            8300   \n",
       "42982  Prof & Tech Engineers - Miscellaneous, Local 21            1000   \n",
       "23310  Prof & Tech Engineers - Miscellaneous, Local 21            5200   \n",
       "...                                                ...             ...   \n",
       "5171   Municipal Executive Association - Miscellaneous            0900   \n",
       "17805  Municipal Executive Association - Miscellaneous            0900   \n",
       "499            Physicians and Dentists - Miscellaneous            2500   \n",
       "13194          Physicians and Dentists - Miscellaneous            2500   \n",
       "19177  Municipal Executive Association - Miscellaneous            1100   \n",
       "\n",
       "                          Job Family  ... Employee Identifier   Salaries  \\\n",
       "27308                  Fire Services  ...               43833       0.00   \n",
       "15746            Public Service Aide  ...               27871    -292.40   \n",
       "24576         Correction & Detention  ...               10517       0.00   \n",
       "42982            Information Systems  ...               18961       0.00   \n",
       "23310       Professional Engineering  ...               19387       0.00   \n",
       "...                              ...  ...                 ...        ...   \n",
       "5171                      Management  ...                1523  256098.01   \n",
       "17805                     Management  ...               17356  326764.01   \n",
       "499          Med Therapy & Auxiliary  ...               13746  279311.03   \n",
       "13194        Med Therapy & Auxiliary  ...                1016  279311.10   \n",
       "19177  Administrative & Mgmt (Unrep)  ...               46881  507831.60   \n",
       "\n",
       "       Overtime  Other Salaries  Total Salary  Retirement  Health/Dental  \\\n",
       "27308      0.00            0.00          0.00        0.00           0.00   \n",
       "15746      0.00            0.00       -292.40        0.00         -95.58   \n",
       "24576      0.00            0.00          0.00        0.00           0.00   \n",
       "42982      0.00            0.00          0.00        0.00           0.00   \n",
       "23310      0.00            0.00          0.00        0.00           0.00   \n",
       "...         ...             ...           ...         ...            ...   \n",
       "5171       0.00        82292.31     338390.32    51977.53       11468.77   \n",
       "17805      0.00            0.00     326764.01    65806.33       12424.50   \n",
       "499     9046.92        56742.56     345100.51    56211.61       12424.50   \n",
       "13194   3829.36       114433.58     397574.04    56211.64       12424.50   \n",
       "19177      0.00            0.00     507831.60   105052.98       12424.50   \n",
       "\n",
       "       Other Benefits  Total Benefits  Total Compensation  \n",
       "27308         -423.76         -423.76             -423.76  \n",
       "15746          -22.63         -118.21             -410.61  \n",
       "24576         -159.12         -159.12             -159.12  \n",
       "42982          -26.53          -26.53              -26.53  \n",
       "23310           -9.51           -9.51               -9.51  \n",
       "...               ...             ...                 ...  \n",
       "5171         20963.32        84409.62           422799.94  \n",
       "17805        21691.23        99922.06           426686.07  \n",
       "499          13482.66        82118.77           427219.28  \n",
       "13194        14299.10        82935.24           480509.28  \n",
       "19177        23566.16       141043.64           648875.24  \n",
       "\n",
       "[42989 rows x 22 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sf2015.sort_values(by=['Total Compensation'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For clarity of comparison, we will focus our attention on those who had at least the equivalent of a half-time job for the whole year. At a minimum wage of about \\$10 per hour, and 20 hours per week for 52 weeks, that's a salary of about \\$10,000."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "sf2015 = sf2015[sf2015['Salaries'] > 10000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36569"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(sf2015)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Population and Parameter\n",
    "Let this table of just over 36,500 rows be our population. Here is a histogram of the total compensations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sf_bins = np.arange(0, 700000, 25000)\n",
    "\n",
    "unit = ''\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,5))\n",
    "\n",
    "ax.hist(sf2015['Total Compensation'], bins = sf_bins, density=True, color='blue', alpha=0.8, ec='white')\n",
    "\n",
    "#ax.scatter(observed_distance, 0, color='red', s=40, zorder=10).set_clip_on(False)\n",
    "\n",
    "y_vals = ax.get_yticks()\n",
    "\n",
    "y_label = 'Percent per ' + (unit if unit else 'unit')\n",
    "\n",
    "x_label = 'Total Compensation'\n",
    "\n",
    "ax.set_yticklabels(['{:g}'.format(x * 100) for x in y_vals])\n",
    "\n",
    "plt.ylabel(y_label)\n",
    "\n",
    "plt.xlabel(x_label)\n",
    "\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "plt.title('');\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While most of the values are below \\\\$300,000,  a few are quite a bit higher. For example, the total compensation of the Chief Investment Officer was almost \\\\$650,000. That is why the horizontal axis stretches to \\\\$700,000."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>Year Type</th>\n",
       "      <th>Year</th>\n",
       "      <th>Organization Group Code</th>\n",
       "      <th>Organization Group</th>\n",
       "      <th>Department Code</th>\n",
       "      <th>Department</th>\n",
       "      <th>Union Code</th>\n",
       "      <th>Union</th>\n",
       "      <th>Job Family Code</th>\n",
       "      <th>Job Family</th>\n",
       "      <th>...</th>\n",
       "      <th>Employee Identifier</th>\n",
       "      <th>Salaries</th>\n",
       "      <th>Overtime</th>\n",
       "      <th>Other Salaries</th>\n",
       "      <th>Total Salary</th>\n",
       "      <th>Retirement</th>\n",
       "      <th>Health/Dental</th>\n",
       "      <th>Other Benefits</th>\n",
       "      <th>Total Benefits</th>\n",
       "      <th>Total Compensation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19177</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>RET</td>\n",
       "      <td>Retirement System</td>\n",
       "      <td>351.0</td>\n",
       "      <td>Municipal Executive Association - Miscellaneous</td>\n",
       "      <td>1100</td>\n",
       "      <td>Administrative &amp; Mgmt (Unrep)</td>\n",
       "      <td>...</td>\n",
       "      <td>46881</td>\n",
       "      <td>507831.6</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>507831.60</td>\n",
       "      <td>105052.98</td>\n",
       "      <td>12424.5</td>\n",
       "      <td>23566.16</td>\n",
       "      <td>141043.64</td>\n",
       "      <td>648875.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13194</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>General Administration &amp; Finance</td>\n",
       "      <td>ADM</td>\n",
       "      <td>General Services Agency - City Admin</td>\n",
       "      <td>164.0</td>\n",
       "      <td>Physicians and Dentists - Miscellaneous</td>\n",
       "      <td>2500</td>\n",
       "      <td>Med Therapy &amp; Auxiliary</td>\n",
       "      <td>...</td>\n",
       "      <td>1016</td>\n",
       "      <td>279311.1</td>\n",
       "      <td>3829.36</td>\n",
       "      <td>114433.58</td>\n",
       "      <td>397574.04</td>\n",
       "      <td>56211.64</td>\n",
       "      <td>12424.5</td>\n",
       "      <td>14299.10</td>\n",
       "      <td>82935.24</td>\n",
       "      <td>480509.28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Year Type  Year  Organization Group Code  \\\n",
       "19177  Calendar  2015                        6   \n",
       "13194  Calendar  2015                        6   \n",
       "\n",
       "                     Organization Group Department Code  \\\n",
       "19177  General Administration & Finance             RET   \n",
       "13194  General Administration & Finance             ADM   \n",
       "\n",
       "                                 Department  Union Code  \\\n",
       "19177                     Retirement System       351.0   \n",
       "13194  General Services Agency - City Admin       164.0   \n",
       "\n",
       "                                                 Union Job Family Code  \\\n",
       "19177  Municipal Executive Association - Miscellaneous            1100   \n",
       "13194          Physicians and Dentists - Miscellaneous            2500   \n",
       "\n",
       "                          Job Family  ... Employee Identifier  Salaries  \\\n",
       "19177  Administrative & Mgmt (Unrep)  ...               46881  507831.6   \n",
       "13194        Med Therapy & Auxiliary  ...                1016  279311.1   \n",
       "\n",
       "       Overtime  Other Salaries  Total Salary  Retirement  Health/Dental  \\\n",
       "19177      0.00            0.00     507831.60   105052.98        12424.5   \n",
       "13194   3829.36       114433.58     397574.04    56211.64        12424.5   \n",
       "\n",
       "       Other Benefits  Total Benefits  Total Compensation  \n",
       "19177        23566.16       141043.64           648875.24  \n",
       "13194        14299.10        82935.24           480509.28  \n",
       "\n",
       "[2 rows x 22 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sf2015.sort_values(by=['Total Compensation'], ascending=False).head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let the parameter be the median of the total compensations.\n",
    "\n",
    "Since we have the luxury of having all of the data from the population, we can simply calculate the parameter:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "110305.79"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_median = np.percentile(sf2015['Total Compensation'], 50)\n",
    "pop_median"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The median total compensation of all employees was just over \\$110,300. \n",
    "\n",
    "From a practical perspective, there is no reason for us to draw a sample to estimate this parameter since we simply know its value. But in this section we are going to pretend we don't know the value, and see how well we can estimate it based on a random sample. \n",
    "\n",
    "In later sections, we will come down to earth and work in situations where the parameter is unknown. For now, we are all-knowing."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A Random Sample and an Estimate\n",
    "Let us draw a sample of 500 employees at random without replacement, and let the median total compensation of the sampled employees serve as our estimate of the parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "our_sample = sf2015.sample(500, replace=False)\n",
    "\n",
    "unit = ''\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,5))\n",
    "\n",
    "ax.hist(our_sample['Total Compensation'], bins = sf_bins, density=True, color='blue', alpha=0.8, ec='white')\n",
    "\n",
    "y_vals = ax.get_yticks()\n",
    "\n",
    "y_label = 'Percent per ' + (unit if unit else 'unit')\n",
    "\n",
    "x_label = 'Total Compensation'\n",
    "\n",
    "ax.set_yticklabels(['{:g}'.format(x * 100) for x in y_vals])\n",
    "\n",
    "plt.ylabel(y_label)\n",
    "\n",
    "plt.xlabel(x_label)\n",
    "\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "plt.title('');\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "106524.32"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "est_median = np.percentile(our_sample['Total Compensation'], 50)\n",
    "est_median"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The sample size is large. By the law of averages, the distribution of the sample resembles that of the population, and consequently the sample median is not very far from the population median (though of course it is not exactly the same).\n",
    "\n",
    "So now we have one estimate of the parameter. But had the sample come out differently, the estimate would have had a different value. We would like to be able to quantify the amount by which the estimate could vary across samples. That measure of variability will help us measure how accurately we can estimate the parameter.\n",
    "\n",
    "To see how different the estimate would be if the sample had come out differently, we could just draw another sample from the population, but that would be cheating. We are trying to mimic real life, in which we won't have all the population data at hand.\n",
    "\n",
    "Somehow, we have to get another random sample without sampling from the population."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The Bootstrap: Resampling from the Sample\n",
    "What we do have is a large random sample from the population. As we know, a large random sample is likely to resemble the population from which it is drawn. This observation allows data scientists to *lift themselves up by their own bootstraps*: the sampling procedure can be replicated by *sampling from the sample*. \n",
    "\n",
    "Here are the steps of *the bootstrap method* for generating another random sample that resembles the population:\n",
    "\n",
    "- **Treat the original sample as if it were the population.**\n",
    "- **Draw from the sample**, at random **with** replacement, **the same number of times as the original sample size**. \n",
    "\n",
    "It is important to resample the same number of times as the original sample size. The reason is that the variability of an estimate depends on the size of the sample. Since our original sample consisted of 500 employees, our sample median was based on 500 values. To see how different the sample could have been, we have to compare it to the median of other samples of size 500.\n",
    "\n",
    "If we drew 500 times at random *without* replacement from our sample of size 500, we would just get the same sample back. By drawing *with* replacement, we create the possibility for the new samples to be different from the original, because some employees might be drawn more than once and others not at all.\n",
    "\n",
    "Why is this a good idea? By the law of averages, the distribution of the original sample is likely to resemble the population, and the distributions of all the \"resamples\" are likely to resemble the original sample. So the distributions of all the resamples are likely to resemble the population as well. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": [
     "remove_input"
    ]
   },
   "source": [
    "![Bootstrap](../../images/bootstrap_pic.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### A Resampled Median\n",
    "Recall that when the `sample` method is used without specifying a sample size, by default the sample size equals the number of rows of the table from which the sample is drawn. That's perfect for the bootstrap! Here is one new sample drawn from the original sample, and the corresponding sample median."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>Year Type</th>\n",
       "      <th>Year</th>\n",
       "      <th>Organization Group Code</th>\n",
       "      <th>Organization Group</th>\n",
       "      <th>Department Code</th>\n",
       "      <th>Department</th>\n",
       "      <th>Union Code</th>\n",
       "      <th>Union</th>\n",
       "      <th>Job Family Code</th>\n",
       "      <th>Job Family</th>\n",
       "      <th>...</th>\n",
       "      <th>Employee Identifier</th>\n",
       "      <th>Salaries</th>\n",
       "      <th>Overtime</th>\n",
       "      <th>Other Salaries</th>\n",
       "      <th>Total Salary</th>\n",
       "      <th>Retirement</th>\n",
       "      <th>Health/Dental</th>\n",
       "      <th>Other Benefits</th>\n",
       "      <th>Total Benefits</th>\n",
       "      <th>Total Compensation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8264</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>250.0</td>\n",
       "      <td>SEIU - Health Workers, Local 1021</td>\n",
       "      <td>2700</td>\n",
       "      <td>Housekeeping &amp; Laundry</td>\n",
       "      <td>...</td>\n",
       "      <td>2712</td>\n",
       "      <td>56531.01</td>\n",
       "      <td>621.29</td>\n",
       "      <td>2350.53</td>\n",
       "      <td>59502.83</td>\n",
       "      <td>11681.45</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>4730.12</td>\n",
       "      <td>28836.07</td>\n",
       "      <td>88338.90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4685</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>250.0</td>\n",
       "      <td>SEIU - Health Workers, Local 1021</td>\n",
       "      <td>2300</td>\n",
       "      <td>Nursing</td>\n",
       "      <td>...</td>\n",
       "      <td>35035</td>\n",
       "      <td>17258.19</td>\n",
       "      <td>4907.16</td>\n",
       "      <td>1878.03</td>\n",
       "      <td>24043.38</td>\n",
       "      <td>4554.88</td>\n",
       "      <td>4696.52</td>\n",
       "      <td>1919.13</td>\n",
       "      <td>11170.53</td>\n",
       "      <td>35213.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12319</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>Public Works, Transportation &amp; Commerce</td>\n",
       "      <td>DPW</td>\n",
       "      <td>General Services Agency - Public Works</td>\n",
       "      <td>21.0</td>\n",
       "      <td>Prof &amp; Tech Engineers - Miscellaneous, Local 21</td>\n",
       "      <td>5200</td>\n",
       "      <td>Professional Engineering</td>\n",
       "      <td>...</td>\n",
       "      <td>4931</td>\n",
       "      <td>135421.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>135421.01</td>\n",
       "      <td>27253.50</td>\n",
       "      <td>12424.52</td>\n",
       "      <td>10116.83</td>\n",
       "      <td>49794.85</td>\n",
       "      <td>185215.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23490</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>3</td>\n",
       "      <td>Human Welfare &amp; Neighborhood Development</td>\n",
       "      <td>DSS</td>\n",
       "      <td>Human Services</td>\n",
       "      <td>21.0</td>\n",
       "      <td>Prof &amp; Tech Engineers - Miscellaneous, Local 21</td>\n",
       "      <td>1200</td>\n",
       "      <td>Personnel</td>\n",
       "      <td>...</td>\n",
       "      <td>43280</td>\n",
       "      <td>92085.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>92085.00</td>\n",
       "      <td>18979.21</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>7567.93</td>\n",
       "      <td>38971.64</td>\n",
       "      <td>131056.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15708</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>790.0</td>\n",
       "      <td>SEIU - Miscellaneous, Local 1021</td>\n",
       "      <td>9900</td>\n",
       "      <td>Public Service Aide</td>\n",
       "      <td>...</td>\n",
       "      <td>35494</td>\n",
       "      <td>37494.70</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>37494.70</td>\n",
       "      <td>8999.16</td>\n",
       "      <td>12177.75</td>\n",
       "      <td>3055.99</td>\n",
       "      <td>24232.90</td>\n",
       "      <td>61727.60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23644</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>5</td>\n",
       "      <td>Culture &amp; Recreation</td>\n",
       "      <td>FAM</td>\n",
       "      <td>Fine Arts Museum</td>\n",
       "      <td>351.0</td>\n",
       "      <td>Municipal Executive Association - Miscellaneous</td>\n",
       "      <td>0900</td>\n",
       "      <td>Management</td>\n",
       "      <td>...</td>\n",
       "      <td>35997</td>\n",
       "      <td>118427.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>118427.00</td>\n",
       "      <td>23833.74</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>24506.58</td>\n",
       "      <td>60764.82</td>\n",
       "      <td>179191.82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8710</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>791.0</td>\n",
       "      <td>SEIU - Staff and Per Diem Nurses, Local 1021</td>\n",
       "      <td>2300</td>\n",
       "      <td>Nursing</td>\n",
       "      <td>...</td>\n",
       "      <td>5225</td>\n",
       "      <td>17916.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>17916.00</td>\n",
       "      <td>3248.16</td>\n",
       "      <td>1911.46</td>\n",
       "      <td>1416.06</td>\n",
       "      <td>6575.68</td>\n",
       "      <td>24491.68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25486</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>Public Protection</td>\n",
       "      <td>POL</td>\n",
       "      <td>Police</td>\n",
       "      <td>911.0</td>\n",
       "      <td>Police Officers' Association</td>\n",
       "      <td>Q000</td>\n",
       "      <td>Police Services</td>\n",
       "      <td>...</td>\n",
       "      <td>32237</td>\n",
       "      <td>140578.74</td>\n",
       "      <td>95.73</td>\n",
       "      <td>10759.56</td>\n",
       "      <td>151434.03</td>\n",
       "      <td>28046.17</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>2569.95</td>\n",
       "      <td>43040.62</td>\n",
       "      <td>194474.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11368</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>Public Protection</td>\n",
       "      <td>CRT</td>\n",
       "      <td>Superior Court</td>\n",
       "      <td>195.0</td>\n",
       "      <td>Court Unrepresented Professionals</td>\n",
       "      <td>SCRT</td>\n",
       "      <td>SF Superior Court</td>\n",
       "      <td>...</td>\n",
       "      <td>43867</td>\n",
       "      <td>68656.96</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3000.00</td>\n",
       "      <td>71656.96</td>\n",
       "      <td>13250.87</td>\n",
       "      <td>7778.27</td>\n",
       "      <td>5949.31</td>\n",
       "      <td>26978.45</td>\n",
       "      <td>98635.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25489</th>\n",
       "      <td>Calendar</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>Community Health</td>\n",
       "      <td>DPH</td>\n",
       "      <td>Public Health</td>\n",
       "      <td>856.0</td>\n",
       "      <td>Teamsters - Miscellaneous, Local 856</td>\n",
       "      <td>2400</td>\n",
       "      <td>Lab, Pharmacy &amp; Med Techs</td>\n",
       "      <td>...</td>\n",
       "      <td>27976</td>\n",
       "      <td>107416.03</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>107416.03</td>\n",
       "      <td>22139.02</td>\n",
       "      <td>12424.50</td>\n",
       "      <td>8208.83</td>\n",
       "      <td>42772.35</td>\n",
       "      <td>150188.38</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>500 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Year Type  Year  Organization Group Code  \\\n",
       "8264   Calendar  2015                        4   \n",
       "4685   Calendar  2015                        4   \n",
       "12319  Calendar  2015                        2   \n",
       "23490  Calendar  2015                        3   \n",
       "15708  Calendar  2015                        4   \n",
       "...         ...   ...                      ...   \n",
       "23644  Calendar  2015                        5   \n",
       "8710   Calendar  2015                        4   \n",
       "25486  Calendar  2015                        1   \n",
       "11368  Calendar  2015                        1   \n",
       "25489  Calendar  2015                        4   \n",
       "\n",
       "                             Organization Group Department Code  \\\n",
       "8264                           Community Health             DPH   \n",
       "4685                           Community Health             DPH   \n",
       "12319   Public Works, Transportation & Commerce             DPW   \n",
       "23490  Human Welfare & Neighborhood Development             DSS   \n",
       "15708                          Community Health             DPH   \n",
       "...                                         ...             ...   \n",
       "23644                      Culture & Recreation             FAM   \n",
       "8710                           Community Health             DPH   \n",
       "25486                         Public Protection             POL   \n",
       "11368                         Public Protection             CRT   \n",
       "25489                          Community Health             DPH   \n",
       "\n",
       "                                   Department  Union Code  \\\n",
       "8264                            Public Health       250.0   \n",
       "4685                            Public Health       250.0   \n",
       "12319  General Services Agency - Public Works        21.0   \n",
       "23490                          Human Services        21.0   \n",
       "15708                           Public Health       790.0   \n",
       "...                                       ...         ...   \n",
       "23644                        Fine Arts Museum       351.0   \n",
       "8710                            Public Health       791.0   \n",
       "25486                                  Police       911.0   \n",
       "11368                          Superior Court       195.0   \n",
       "25489                           Public Health       856.0   \n",
       "\n",
       "                                                 Union Job Family Code  \\\n",
       "8264                 SEIU - Health Workers, Local 1021            2700   \n",
       "4685                 SEIU - Health Workers, Local 1021            2300   \n",
       "12319  Prof & Tech Engineers - Miscellaneous, Local 21            5200   \n",
       "23490  Prof & Tech Engineers - Miscellaneous, Local 21            1200   \n",
       "15708                 SEIU - Miscellaneous, Local 1021            9900   \n",
       "...                                                ...             ...   \n",
       "23644  Municipal Executive Association - Miscellaneous            0900   \n",
       "8710      SEIU - Staff and Per Diem Nurses, Local 1021            2300   \n",
       "25486                     Police Officers' Association            Q000   \n",
       "11368                Court Unrepresented Professionals            SCRT   \n",
       "25489             Teamsters - Miscellaneous, Local 856            2400   \n",
       "\n",
       "                      Job Family  ... Employee Identifier   Salaries  \\\n",
       "8264      Housekeeping & Laundry  ...                2712   56531.01   \n",
       "4685                     Nursing  ...               35035   17258.19   \n",
       "12319   Professional Engineering  ...                4931  135421.01   \n",
       "23490                  Personnel  ...               43280   92085.00   \n",
       "15708        Public Service Aide  ...               35494   37494.70   \n",
       "...                          ...  ...                 ...        ...   \n",
       "23644                 Management  ...               35997  118427.00   \n",
       "8710                     Nursing  ...                5225   17916.00   \n",
       "25486            Police Services  ...               32237  140578.74   \n",
       "11368          SF Superior Court  ...               43867   68656.96   \n",
       "25489  Lab, Pharmacy & Med Techs  ...               27976  107416.03   \n",
       "\n",
       "       Overtime  Other Salaries  Total Salary  Retirement  Health/Dental  \\\n",
       "8264     621.29         2350.53      59502.83    11681.45       12424.50   \n",
       "4685    4907.16         1878.03      24043.38     4554.88        4696.52   \n",
       "12319      0.00            0.00     135421.01    27253.50       12424.52   \n",
       "23490      0.00            0.00      92085.00    18979.21       12424.50   \n",
       "15708      0.00            0.00      37494.70     8999.16       12177.75   \n",
       "...         ...             ...           ...         ...            ...   \n",
       "23644      0.00            0.00     118427.00    23833.74       12424.50   \n",
       "8710       0.00            0.00      17916.00     3248.16        1911.46   \n",
       "25486     95.73        10759.56     151434.03    28046.17       12424.50   \n",
       "11368      0.00         3000.00      71656.96    13250.87        7778.27   \n",
       "25489      0.00            0.00     107416.03    22139.02       12424.50   \n",
       "\n",
       "       Other Benefits  Total Benefits  Total Compensation  \n",
       "8264          4730.12        28836.07            88338.90  \n",
       "4685          1919.13        11170.53            35213.91  \n",
       "12319        10116.83        49794.85           185215.86  \n",
       "23490         7567.93        38971.64           131056.64  \n",
       "15708         3055.99        24232.90            61727.60  \n",
       "...               ...             ...                 ...  \n",
       "23644        24506.58        60764.82           179191.82  \n",
       "8710          1416.06         6575.68            24491.68  \n",
       "25486         2569.95        43040.62           194474.65  \n",
       "11368         5949.31        26978.45            98635.41  \n",
       "25489         8208.83        42772.35           150188.38  \n",
       "\n",
       "[500 rows x 22 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resample_1 = our_sample.sample(len(our_sample), replace=True)\n",
    "resample_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "unit = ''\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,5))\n",
    "\n",
    "ax.hist(resample_1['Total Compensation'], bins = sf_bins, density=True, color='blue', alpha=0.8, ec='white')\n",
    "\n",
    "y_vals = ax.get_yticks()\n",
    "\n",
    "y_label = 'Percent per ' + (unit if unit else 'unit')\n",
    "\n",
    "x_label = 'Total Compensation'\n",
    "\n",
    "ax.set_yticklabels(['{:g}'.format(x * 100) for x in y_vals])\n",
    "\n",
    "plt.ylabel(y_label)\n",
    "\n",
    "plt.xlabel(x_label)\n",
    "\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "plt.title('');\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "111355.89"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resampled_median_1 = np.percentile(resample_1['Total Compensation'], 50, interpolation='nearest')\n",
    "resampled_median_1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By resampling, we have another estimate of the population median. By resampling again and again, we will get many such estimates, and hence an empirical distribution of the estimates."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "107778.0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resample_2 = our_sample.sample(len(our_sample), replace=True)\n",
    "resampled_median_2 = np.percentile(resample_2['Total Compensation'], 50, interpolation='nearest')\n",
    "resampled_median_2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bootstrap Empirical Distribution of the Sample Median\n",
    "Let us define a function `bootstrap_median` that takes our original sample, the label of the column containing the variable, and the number of bootstrap samples we want to take, and returns an array of the corresponding resampled medians. \n",
    "\n",
    "Each time we resample and find the median, we *replicate* the bootstrap process. So the number of bootstrap samples will be called the number of replications."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bootstrap_median(original_sample, label, replications):\n",
    "    \"\"\"Returns an array of bootstrapped sample medians:\n",
    "    original_sample: table containing the original sample\n",
    "    label: label of column containing the variable\n",
    "    replications: number of bootstrap samples\n",
    "    \"\"\"\n",
    "    just_one_column = original_sample[label]\n",
    "    medians = np.array([])\n",
    "    for i in np.arange(replications):\n",
    "        bootstrap_sample = just_one_column.sample(len(just_one_column), replace=True)\n",
    "        resampled_median = np.percentile(bootstrap_sample, 50)\n",
    "        medians = np.append(medians, resampled_median)\n",
    "        \n",
    "    return medians"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now replicate the bootstrap process 5,000 times. The array `bstrap_medians` contains the medians of all 5,000 bootstrap samples. Notice that the code takes longer to run than our previous code. It has a lot of resampling to do!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "bstrap_medians = bootstrap_median(our_sample, 'Total Compensation', 5000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is the histogram of the 5000 medians. The red dot is the population parameter: it is the median of the entire population, which we happen to know but did not use in the bootstrap process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "resampled_medians = pd.DataFrame({'Bootstrap Sample Median':bstrap_medians})\n",
    "\n",
    "unit = ''\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,5))\n",
    "\n",
    "ax.hist(resampled_medians, density=True, color='blue', alpha=0.8, ec='white')\n",
    "\n",
    "ax.scatter(pop_median, 0, color='red', s=40, zorder=10).set_clip_on(False)\n",
    "\n",
    "y_vals = ax.get_yticks()\n",
    "\n",
    "y_label = 'Percent per ' + (unit if unit else 'unit')\n",
    "\n",
    "x_label = 'Bootstrap Sample Median'\n",
    "\n",
    "ax.set_yticklabels(['{:g}'.format(x * 100) for x in y_vals])\n",
    "\n",
    "plt.ylabel(y_label)\n",
    "\n",
    "plt.xlabel(x_label)\n",
    "\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "plt.title('');\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is important to remember that the red dot is fixed: it is \\$110,305.79, the population median. The empirical histogram is the result of random draws, and will be situated randomly relative to the red dot. \n",
    "\n",
    "Remember also that the point of all these computations is to estimate the population median, which is the red dot. Our estimates are all the randomly generated sampled medians whose histogram you see above. We want those estimates to contain the parameter – it they don't, then they are off."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Do the Estimates Capture the Parameter?\n",
    "How often does the empirical histogram of the resampled medians sit firmly over the red dot, and not just brush the dot with its tails? To answer this, we must define \"sit firmly\". Let's take that to mean \"the middle 95% of the resampled medians contains the red dot\". \n",
    "\n",
    "Here are the two ends of the \"middle 95%\" interval of resampled medians:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "98658.98000000001"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left = np.percentile(bstrap_medians, 2.5, interpolation='nearest')\n",
    "left"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "114416.98999999999"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "right = np.percentile(bstrap_medians, 97.5, interpolation='nearest')\n",
    "right"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The population median of \\$110,305 is between these two numbers. The interval and the population median are shown on the histogram below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "resampled_medians = pd.DataFrame({'Bootstrap Sample Median':bstrap_medians})\n",
    "\n",
    "unit = ''\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,5))\n",
    "\n",
    "ax.hist(resampled_medians, density=True, color='blue', alpha=0.8, ec='white', zorder=5)\n",
    "\n",
    "ax.plot(np.array([left, right]), np.array([0,0]), color='yellow', lw=8, zorder=10)\n",
    "\n",
    "ax.scatter(pop_median, 0, color='red', s=40, zorder=15).set_clip_on(False)\n",
    "\n",
    "y_vals = ax.get_yticks()\n",
    "\n",
    "y_label = 'Percent per ' + (unit if unit else 'unit')\n",
    "\n",
    "x_label = 'Bootstrap Sample Median'\n",
    "\n",
    "ax.set_yticklabels(['{:g}'.format(x * 100) for x in y_vals])\n",
    "\n",
    "plt.ylabel(y_label)\n",
    "\n",
    "plt.xlabel(x_label)\n",
    "\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "plt.title('');\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The \"middle 95%\" interval of estimates captured the parameter in our example. But was that a fluke? \n",
    "\n",
    "To see how frequently the interval contains the parameter, we have to run the entire process over and over again. Specifically, we will repeat the following process 100 times:\n",
    "\n",
    "- Draw an original sample of size 500 from the population.\n",
    "- Carry out 5,000 replications of the bootstrap process and generate the \"middle 95%\" interval of resampled medians.\n",
    "\n",
    "We will end up with 100 intervals, and count how many of them contain the population median.\n",
    "\n",
    "**Spoiler alert:** The statistical theory of the bootstrap says that the number should be around 95. It may be in the low 90s or high 90s, but not much farther off 95 than that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Total Compensation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>117766.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>41209.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>110561.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>38624.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>260280.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42983</th>\n",
       "      <td>61349.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42984</th>\n",
       "      <td>132788.81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42986</th>\n",
       "      <td>73295.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42987</th>\n",
       "      <td>19973.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42988</th>\n",
       "      <td>55812.90</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>36569 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Total Compensation\n",
       "0               117766.86\n",
       "1                41209.83\n",
       "2               110561.13\n",
       "3                38624.97\n",
       "6               260280.95\n",
       "...                   ...\n",
       "42983            61349.71\n",
       "42984           132788.81\n",
       "42986            73295.59\n",
       "42987            19973.37\n",
       "42988            55812.90\n",
       "\n",
       "[36569 rows x 1 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_comps = sf2015[['Total Compensation']]\n",
    "total_comps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# THE BIG SIMULATION: This one takes several minutes.\n",
    "\n",
    "# Generate 100 intervals, in the table intervals\n",
    "\n",
    "left_ends = np.array([])\n",
    "right_ends = np.array([])\n",
    "\n",
    "total_comps = sf2015[['Total Compensation']]\n",
    "\n",
    "for i in np.arange(100):\n",
    "    first_sample = total_comps.sample(500, replace=False)\n",
    "    medians = bootstrap_median(first_sample, 'Total Compensation', 5000)\n",
    "    left_ends = np.append(left_ends, np.percentile(medians, 2.5))\n",
    "    right_ends = np.append(right_ends, np.percentile(medians, 97.5))\n",
    "\n",
    "intervals = pd.DataFrame(\n",
    "    {'Left':left_ends,\n",
    "    'Right':right_ends}\n",
    ")    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each of the 100 replications, we get one interval of estimates of the median."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>Left</th>\n",
       "      <th>Right</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>100639.945000</td>\n",
       "      <td>113009.80850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>103596.780000</td>\n",
       "      <td>113549.15000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>106650.085125</td>\n",
       "      <td>116872.21000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>101855.280000</td>\n",
       "      <td>111880.11000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>98633.755000</td>\n",
       "      <td>113396.61500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>103042.430000</td>\n",
       "      <td>117401.49500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>97683.995000</td>\n",
       "      <td>113182.87500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>111704.995000</td>\n",
       "      <td>121917.01000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>105761.655000</td>\n",
       "      <td>117376.24225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>104795.530000</td>\n",
       "      <td>116920.50500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Left         Right\n",
       "0   100639.945000  113009.80850\n",
       "1   103596.780000  113549.15000\n",
       "2   106650.085125  116872.21000\n",
       "3   101855.280000  111880.11000\n",
       "4    98633.755000  113396.61500\n",
       "..            ...           ...\n",
       "95  103042.430000  117401.49500\n",
       "96   97683.995000  113182.87500\n",
       "97  111704.995000  121917.01000\n",
       "98  105761.655000  117376.24225\n",
       "99  104795.530000  116920.50500\n",
       "\n",
       "[100 rows x 2 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intervals"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The good intervals are those that contain the parameter we are trying to estimate. Typically the parameter is unknown, but in this section we happen to know what the parameter is."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "110305.79"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_median"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How many of the 100 intervals contain the population median? That's the number of intervals where the left end is below the population median and the right end is above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "95"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(\n",
    "    intervals[\n",
    "        (intervals['Left'] < pop_median) & \n",
    "              (intervals['Right'] > pop_median)\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It takes a few minutes to construct all the intervals, but try it again if you have the patience. Most likely, about 95 of the 100 intervals will be good ones: they will contain the parameter.\n",
    "\n",
    "It's hard to show you all the intervals on the horizontal axis as they have large overlaps – after all, they are all trying to estimate the same parameter. The graphic below shows each interval on the same axes by stacking them vertically. The vertical axis is simply the number of the replication from which the interval was generated.\n",
    "\n",
    "The red line is where the parameter is. Good intervals cover the parameter; there are about 95 of these, typically. \n",
    "\n",
    "If an interval doesn't cover the parameter, it's a dud. The duds are the ones where you can see \"daylight\" around the red line. There are very few of them – about 5, typically – but they do happen. \n",
    "\n",
    "Any method based on sampling has the possibility of being off. The beauty of methods based on random sampling is that we can quantify how often they are likely to be off."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAIdCAYAAADBH1z+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAABWCklEQVR4nO3dd3yTVfvH8W8oqyA0lae0rLJahogg24KsIkNAZFcRAQUU8HGCFXkUBLWgIMsKKDhQBJnKFmTJhgeBKq4KgghYZtmb+/cHT/MjTUqTJm3Su5/368VLe98nyZWTk7tXz7lyYklOTjYEAABgErl8HQAAAIA3kdwAAABTIbkBAACmQnIDAABMheQGAACYCskNAAAwFZIbOIiLi5PVatX69esz9XFat24tq9WaqY9hBimvx4wZM+yOV61aVVWrVvVRVP5p/fr1slqt6tevn69DMYUZM2Y4HXu+snfvXj322GOqVKmSgoOD/f76YbVa1bp1a1+HkSOR3HiZ1Wq1+3fnnXeqdOnSatmypT755BNdv37d1yFmmZTk5cCBA74OxWUpiYTValWfPn3SbLdlyxZbuwoVKmRhhPCmAwcOyGq1eiVJTCsJhXdcv35d3bp105IlS9SwYUMNGjRIsbGx6d4u9TXZ2b/FixdnKCZvjZ2slFPGaW5fB2BWKW+669ev688//9TixYu1ZcsWrV27Vp999pmPo/MPkydP1sWLF30dhlO5c+fWwoUL9c477yg4ONjh/GeffabcuXPr2rVrPojupoULF/rssYGsduDAAf3666+Kjo7Whx9+6Pbtb5cIZdYfKNu2bVNgYGCm3Dduj+QmkwwePNju5z179qhZs2b65ptvtGnTJkVFRfkoMv9RqlQpX4eQphYtWmjJkiWaNWuWwxLH6dOn9c0336hly5YZ/ovPG8qWLeuzxway2pEjRyRJRYsWzdDtU1+TswKzur7DslQWqVKliurXry9J2rFjh+14UlKSBg0apGrVqqlo0aIqW7asunTpog0bNjjcx631BL///rseeeQRlSlTRsWLF1erVq20du1ah9ukVz/jzprw4sWL1adPH9WoUUPFixdXiRIl1LBhQ33wwQcOy21Wq1UbN26UJFWrVs02/XvrFG5aNTeGYejTTz9VdHS0SpYsqWLFiqlBgwaaOHGirly54tC+atWqslqtunbtmsaMGaMaNWqoaNGiqlKlil577TVdvnzZped3qyZNmqhUqVKaPn26w7k5c+bowoUL6tGjx23v4/vvv1dMTIzKly+vkJAQ3X333XrppZeUlJTktP2uXbvUsWNHlSxZUqVKlVK7du20devWNO/fWc3N6dOnNX78eLVp00aVK1dWSEiIypcvr5iYmDTvK+V1uXDhgl577TXdfffdKlq0qO69916NHTtWhuH6N7Ts2rVLL7/8sqKiolS6dGmFhoaqRo0aevXVV3Xq1CmH9ik1HXFxcUpISFCXLl0UHh6uYsWKqVWrVtqyZYvTxzl69KieeeYZRUZGKiwsTA0aNPDqNPut77UDBw7oiSeeULly5RQaGqpGjRpp6dKldu1bt26tUaNGSZIGDBhgt+Rx67LsjRs3NH36dLVo0ULh4eEKDQ3Vfffdp/fee8/p2E55bZKTkzVo0CBVqVJFRYoU0QcffKCOHTvKarVq165dTp/Dd999J6vVqr59+9qOufv6pGX37t168sknVbVqVYWGhqpcuXKKiorSSy+9pNOnT7t8P99//706d+6ssmXLqmjRoqpWrZpiY2N17Ngxh35IuU7NnDnT1rdxcXEuP5arLl++rA8++EANGzZUmTJlFBYWprvvvludOnWyzZamjA9JOnjwoN3rfesfQ86ur7cuC61Zs0atWrVSiRIlVL58efXv31/JycmSpJ07d6pz584qXbq0SpQooZiYGKdL/O68ppk1TtevX6+uXbuqSpUqKlq0qCIiItS4cWMNGTLEreuHNzFz40MHDhxQq1atdPjwYdWvX18dOnTQP//8o6+//lrfffedxo0bp8cff9zp7R544AFVqVJFvXr10uHDh/X111+rQ4cO+uSTT9SuXbtMifeNN95Qrly5VKtWLRUvXlynT5/WunXr9Oqrr+qHH37Q1KlTbW1jY2P15Zdf6uDBg3r66acVFBQkSbb/3k7fvn01Z84cFS9eXI8++qjy5Mmj5cuX67XXXtN3332nefPmKXdux6Hbu3dvbd68Wc2aNVOhQoW0cuVKTZw4UceOHdPkyZPdeq65cuVS9+7d9fbbb2v79u2qXbu27dxnn32mkiVLKjo6Os3bjxs3TsOGDVNwcLCaN2+u0NBQ7dmzR9OmTdOyZcu0cuVKlShRwtZ+69atevjhh3X58mW1bdtW5cuX1549e9S2bVs1bNjQ5bh///13vfnmm4qKilKLFi1ktVp18OBBLV26VCtXrtTMmTPVvHlzh9tdu3bNNv6aNWum3Llza8mSJXrjjTd08eJFvfrqqy49/meffabFixerfv36atKkia5fv65du3bpgw8+0MqVK7V69WoVKlTI4Xa7du3ShAkTVLduXT3++OP6+++/tXDhQrVr107ff/+9KlasaGt78uRJNW/eXPv371fdunUVFRWlf/75RwMHDlSjRo1c7itXHDx4UNHR0Spbtqy6du2qU6dOacGCBerWrZu+/vpr2+M9+uijkqSNGzfqwQcftEs6U8b8tWvX9Nhjj2n58uWKiIhQx44dlS9fPm3cuFHDhw/XunXrnI7tK1eu6KGHHtKZM2fUvHlz5c2b1/beWLVqlWbMmKHq1as7xD5z5kxJUrdu3WzHMvr63CohIUHNmzeXxWJRy5YtVbZsWZ07d05//fWXvvzySw0YMMCl9/knn3yiF198UYGBgWrXrp3CwsK0detWTZkyRUuWLNGyZctss7uxsbH666+/NHPmTN199922hKFBgwbpPo67nn76aS1YsECVKlVSly5dVLBgQR05ckQ//PCDFi9erIceekjh4eGKjY3VqFGjVLhwYbuExtUanJTrQKtWrdSzZ0+tW7dOX375pfbv36+hQ4fq4YcfVsOGDdW9e3ft2LFDy5cv1/79+7Vp0yblyvX/8xLuvKaZMU5XrFihrl27qlChQrZELTk5WXv37tWUKVP0xhtvOL1eZ7rk5GSDf977J8mQ5HB88+bNRmBgoCHJWLJkiZGcnGw0bdrUkGS88sordm03btxoBAYGGvny5TN++ukn2/FFixbZ7v/f//633W2+++47IyAgwAgODjb+/vtv2/HY2FhDkrFo0aI0461fv77dsbRus3PnTofbnzx50ujSpYshyVi5cqXdufr16xuSjN27dzt97JTztx6bOnWqIcmoUqWK8ddff9mOHz161GjYsKEhyRg+fLjdbUqVKmVIMu69915j//79tuOHDh0yypYta+TKlcv49ddfXXr9Up77e++9Z/z8889GQECA8dhjj9nOr1mzxu41k2QULVrU7j6WLFliWCwWo3bt2nbxJCcnG5MnTzYkGW3atLEdO3XqlBEZGWlIMj777DO79u+++67tNY+Pj3d43qVKlbI7duDAAWPv3r0OzyshIcEoWrSoERkZmeaYbdGihfHPP//YjicmJhqFCxc2ChcubBw7dsyl/ktISDBOnDjhcHzs2LGGJGPo0KF2x+Pj422PP2XKFKe3eeKJJ+yO9+jRw5Bk9OnTx+74unXrjDx58hiSjEceecSleHfv3m1IcujHW99rQ4YMsTs3b948Q5LRrFkzp2Mn9euU8m/IkCG2uG/to5MnT9qe08iRI52+No0aNTIOHz5sd+6ff/4xgoKCjDvvvNM4evSowzjInz+/UbJkSePkyZMevz63Pqf+/fsbkozPP//c4X4OHjxoJCUluTRO8ubNaxQsWNDYvHmz3bmBAwcakozmzZs7fU1cfW1T92FsbGya/27tN4vFYlSrVs04fvy4w32lfm85Gzupz6d1fc2TJ4/dNfPo0aPGXXfdZUgyChcubNe/p06dMqKjow1JxowZMzx6z3l7nLZp08aQZKxbt87hvvbt2+fWa+XNfyxLZZK4uDjFxcXpzTffVJ8+fdSkSRNdvHhRbdq0Uf369XXo0CGtXr1axYsX14svvmh32ypVquiJJ57Q5cuX9dVXXzncd+HChfXyyy/bHatVq5bat2+vU6dOOUyZe4uzGo9cuXKpf//+kqTVq1d7/BhffPGFJGno0KEqXLiw7XjevHn19ttvS1KaBdnDhg2zW+YqWLCgunTpohs3bqQ5dX87xYsXV7NmzbRgwQKdPXvW9ti5cuXSY489lubtJk+eLMMwNHbsWIdlt5iYGN1zzz1atmyZzpw5I+nmrE1iYqLq1q3rMOv25JNPqly5ci7HHBQUpCJFijgcDw8PV7t27ZSYmKiDBw86ve2oUaOUP39+288hISFq3bq1zpw5o8TERJcePzw8XAEBAQ7He/bsqcKFC6c5Ru677z517drV7thjjz2m3Llz64cffrAdu3r1qubMmaOCBQtqyJAhdu2rVaumLl26uBSnq8LDwx3en9HR0SpVqpRdXOm5ceOGJk+erJCQEMXFxdn1Ua5cuTR8+HBZLBan73dJGjFihAoUKGB3LH/+/OrYsaNOnjyp5cuX251bsGCBLl26pJiYGLu/8jP6+twq5f5SxyNJhQoVUr58+dK9j9mzZ+vKlSt68sknVblyZbtzgwYNUrFixbRixQodPnw43fty1ahRo9L8lyJXrlwyDEP58uVz2k/O3lsZ1blzZ7sZ4bx58+rhhx+WdHMst23b1nbOYrGoc+fOkqQff/zR7n688ZqmyMg4vd14uPPOO11+bG9jWSqTpLxhLBaLChUqpGrVqqlz587q2bOnpJtTu5JUr1495c2b1+H2jRs3Vnx8vHbv3u1wrlq1ak6njuvXr6+5c+faahe87eTJk5owYYJWrFihAwcO6Pz583bnUwr+PJHyfO+//36Hc3fffbdCQkK0d+9enTt3TnfccYfdeWdT8ylLP8n/W8d2V48ePfTtt99q7ty56ty5s+bNm6dmzZqpZMmSad5m69atyp07txYtWqRFixY5nL9y5YquX7+uffv2qXr16rbnnFKTdatcuXKpXr162rdvn8sxb9myRZMnT9b27dt17NgxhzXyI0eOOBRzBwUFqUyZMg735W7/Xb16VZ988onmz5+vX375RWfPntWNGzfsHtsZZ69dnjx5VLRoUbvH/v3333XhwgXVqVPHab1W/fr1vVp7U7VqVae/OEqUKKFt27a5fD9//PGHTpw4obJly+rdd9912iYwMNBpEpkvX740lzoeffRRffzxx/ryyy/10EMP2Y6nLEmlLEOkyOjrc6uOHTtq8uTJ6tatmx566CE1bNhQderUcat4NmXMO1tyzZcvn+rVq6cFCxYoISFBxYsXd/l+b8eVMVyoUCE9+OCDWrp0qerXr682bdrovvvuU+3atR2uN5665557HI6FhYVJcr60lXIudcLnjdc0RUbGaZcuXbRw4UJFR0erffv2uv/++1W7dm2VLl3a5cfNDCQ3mSS9N1LKX+1pVf6HhobatbtVWrcJCQlJ8zaeSk5OVpMmTXTgwAHVrFlTMTExCg4OVkBAgE6fPq3JkydnqHA3tTNnzqhw4cJpfnwyNDRUx44d05kzZxwuNs7W+VN+MWV0f6EWLVqoePHimj59ugICAnT27FmndVC3OnnypK5du2b3F6Ez586dk/T/r1fK65eaO58OWbRokXr06KH8+fOrSZMmKlOmjAoUKKBcuXJpw4YN2rhxo9PX6dZZslu523+9evXS4sWLVaZMGbVu3VqhoaG25H3SpElpjpHbPf6tj+3NvnLF7eK69RdIek6ePClJ+vPPP9MdF6mFhITIYrE4PVerVi1VqlRJ3333nY4dO6aQkBDt27dPW7du1X333ecw65fR1+dW9957r7799luNHj1aixcv1uzZsyXdnEF4/vnn9cQTT6R7H55c/zLbxx9/rIkTJ2rOnDl65513JN1MtFu2bKk333zTa7+0nf2BmvJ+u925q1ev2h33xmuaIiPjtE2bNpo3b54mTpyomTNn2mbW77rrLsXGxmZaDWh6SG58JOWiefToUafnUz5R4+zimtZtUj5hcOttUqYMnf1ycmc24/PPP9eBAwcUGxvr8JHKbdu2uV2wm5bChQvr1KlTunjxotME53b9khkCAgL06KOPavTo0Tp69KjCwsLUsmXL296mcOHCunr1aprLP87aS3L4hEiKtF5vZ95++23lzZtXa9assSvClaTnn3/e9gm2zLBz504tXrxYjRo10ty5c5UnTx7buRs3bmjChAkeP4Y3+yorpcTdsmVLzZo1y63bppXYpHjkkUc0dOhQffXVV3rmmWf05Zdf2o7fypuvT82aNTVz5kxduXJFCQkJWr16tT766CO9+OKLKlCggGJiYm57e0+uf5ktf/78GjRokAYNGqQjR45o8+bNmj17thYtWqRff/1VmzZtsus7X/L2ey6j4zQ6OlrR0dG6ePGiduzYoe+++07Tpk1Tz549tWjRokwp/E4PNTc+kjIluXXrVqcfrVu3bp0k59P1u3fvttWA3CrlF9et050pU/d///23Q/udO3e6HG/KssitU9+pHze1lL803PkLt1q1apLk9KPwP//8s44dO6aIiAivTxHfTvfu3WWxWHTo0CF169Yt3cr/2rVr6+zZsw5r42lJec7O+vHGjRtpfhzamX379qlixYoOiY2795MRKWPkwQcfdLj479ixwysbNlaoUEEFChTQnj17nCbnmZm8ped2s1wVKlRQUFCQduzY4fT97omuXbsqICBAM2fOlGEYmjVrlgoUKKD27dvbtcuM1ydv3ryqVauWXn75ZU2ZMkWSXNr7KWXMO9ui4vLly7ZtC1La+UqxYsXUoUMHzZo1S3Xq1FFiYqJ+++032/lcuXK5dX3ztoy8ppk5TgMDA9WgQQMNGzZMI0aMkGEYmVYDmh6SGx8pUaKEoqOjdejQIY0fP97u3C+//KKPP/5Y+fLlc1o7c+bMGdt0aYr//ve/WrBggaxWqx588EHb8ZSCtS+++MJuOvPEiRN67bXXXI43PDxckuPFaPfu3Ro7dqzT26QU37k6gyHdTCQkafjw4bZlG+nmVGxKAWl6y0LeVrp0aS1YsEBffPGFS99ZNGDAAEk3Z0oOHTrkcP7SpUvavHmz7ee6desqMjJSW7du1TfffGPXdtq0aW7V24SHh2vfvn126/KGYWjkyJH69ddfXb6fjEgZI6kT02PHjmngwIFeeYw8efKoc+fOOn/+vN566y27c7t377YtkfhCynh39odE7ty59fTTT9v64sKFCw5tTpw4YavFc0dYWJiio6O1Z88evf/++/r777/Vpk0bh6UNb70+mzZtcppYpsy23FqUnpYuXboob968mjZtmn7//Xe7c++9954OHz6s5s2bq1ixYi7H5Q3Hjx/X9u3bHY5fvnzZtn/Prc+vSJEiOn78uM92Ws/Ia+rtcbp27Vqn7dwZD5mBZSkfeu+999SyZUu99dZb+v7771W7dm3bPjcXL17U+PHjnRauRkVFafr06dqxY4fq1aunw4cPa8GCBTIMQ+PHj7eb1ahRo4YaNWqkdevWqXHjxmrcuLFOnTqlFStWqFGjRvrpp59cijUmJkYTJkzQq6++qg0bNqh8+fLau3evvv32W7Vt21bz5893uE2TJk20YMECPffcc2rXrp0KFiyooKAgu03FUuvYsaOWL1+uOXPmqF69emrdurVtn5s//vhDjRo18smXIjZu3Njltg0bNtSIESM0dOhQ1axZUw888IDKlCmjS5cu6eDBg9q0aZPCw8NtFySLxaKJEyeqffv26tWrl90+N2vWrFGzZs303XffufTY/fv31wsvvKBGjRrpoYceUu7cubV161b99ttvatmypcOnarypRo0aqlevnhYtWqTmzZurXr16Onr0qL777jtFRkZ67RfV66+/rnXr1umjjz5SQkKCoqKilJSUpAULFqhZs2ZatmyZVx7HXY0aNVKuXLk0efJknTp1ylZP0rdvXwUFBWnQoEH6+eefNX36dK1YsUINGzZUiRIldPz4cf3555/asmWLevfu7bTQND2PPvqoVqxYoeHDh0uy39smhbden/fff1+rV69WgwYNVKZMGRUqVEh//PGHvv32WwUGBrr0/gwPD9eoUaP04osvqkmTJnr44YcVGhqqrVu3auPGjSpRooTGjBnjXiek43Yb/jVr1ky1a9fW4cOH9cADDygyMlLVq1dXiRIldP78ea1evVp79+5V27ZtFRERYbtdkyZNNHv2bHXs2FFRUVHKly+f7r77brVq1cqrsaclI6+pt8fpf/7zH/3111+qX7++wsPDlT9/fu3Zs0erVq3SnXfeme5mp5mF5MaHSpcurbVr12r06NFavny5tmzZooIFC6p+/fp69tlnnX5iKOV248aN09ChQzVt2jRduXJFNWrUUGxsrNNfwp9//rneeOMNLV68WB999JHCw8P173//W88884zmzZvnUqzFihXTsmXLNGzYMG3ZskWrV69WZGSkxowZo0aNGjlNbh577DEdOnRIs2fPVnx8vK5evapSpUrdNrmRpClTpigqKkqff/65Pv/8c924cUPly5fX8OHD9fTTT/vNevft/Pvf/1a9evU0efJkbd68WcuXL9cdd9yhYsWKqXPnzurQoYNd+3r16mnZsmUaMWKEVq1apVWrVqlmzZpavHixVq1a5XJy06tXL+XNm1eTJk3SzJkzlT9/ft13332Kj4/XwoULMzW5SVkaefPNN7VixQpNmTJFxYoV0+OPP66BAweqbt26XnmcIkWK6Ntvv9Xw4cO1fPly7d69WxERERo9erTCw8N9ltxERERo2rRpGj9+vL744gvbX/NdunRRUFCQcufOrenTp2vevHmaMWOGVq5cqXPnzunOO+9UqVKl9MILL6Rbq5KWVq1aKTg4WKdOnVLJkiWdXju89fr07t1bwcHB2rFjh7Zt26arV6+qWLFiiomJ0TPPPOPyp6Z69eqlcuXKaeLEiVqyZInOnz+vYsWKqW/fvho4cKDXi8NvVyAbFBSk2rVrKzw8XK+++qrWr1+vjRs36vjx4woKClK5cuX03HPPOXz6bOTIkcqVK5fWrFmjrVu36vr163rkkUeyLLnJyGvq7XH60ksvacmSJdq5c6dtZr948eLq16+f+vfvf9tPlmYmy/82GkI2sH79erVt21aPPPKIJk2a5OtwAADwS9TcAAAAUyG5AQAApkJyAwAATIWaGwAAYCrM3AAAAFMhuQEAAKZCcgMAAEyF5MbEbv1aemQMfegZ+s8z9J/n6EPPZNf+I7kBAACmQnIDAABMheQGAACYCskNAAAwFZIbAABgKiQ3AADAVEhuAACAqZDcAAAAUyG5AQAApkJyAwAATIXkBgAAmArJDQAAMBWSGwAAYCokNwAAwFR8mtxs3LhRMTExqly5sqxWq2bMmGF33jAMxcXFqVKlSgoLC1Pr1q31yy+/2LW5fPmyBg0apHLlyql48eKKiYnRoUOHsvJpAAAAP+LT5Ob8+fO66667NHLkSAUGBjqcHz9+vOLj4zVq1CitXr1aISEhat++vc6ePWtrM3jwYC1atEjTpk3T0qVLdfbsWXXt2lXXr1/PyqcCAAD8hCU5OdnwdRCSVKJECb3zzjvq1q2bpJuzNpUqVVKfPn00cOBASdLFixcVGRmpESNGqFevXjp9+rQiIiIUHx+vLl26SJL+/vtvVa1aVXPnzlV0dLTPno8/SExMVGRkpK/DyFaCjlh9HQLgF04XS/Z1CF7BddAz2bX//Lbm5sCBA0pKSlLTpk1txwIDAxUVFaWtW7dKknbt2qWrV6/atSlZsqQqVqxoawMAAHKW3L4OIC1JSUmSpJCQELvjISEhOnLkiCTp6NGjCggIUJEiRRzaHD16NM37TkxM9HK0/isnPVdvqHWHryMwucr/++8vt20FP2Cma4eZnosv+GP/pTeb5LfJTQqLxWL3s2EYDsdSS69Ndpxiy4jsOp3oU0d8HQDgH8xy7eA66Jns2n9+m9yEhoZKujk7U7JkSdvx48eP22ZzihYtquvXr+vEiRP617/+ZdcmKioqawOGKaSuM3D3jU3NDszidmPZLPU4MC+/rbkpXbq0QkNDtWbNGtuxS5cuafPmzapbt64kqXr16sqTJ49dm0OHDum3336ztQEAADmLT2duzp07p3379kmSbty4ob///lsJCQkKDg5WqVKl1K9fP40ZM0aRkZGKiIjQ6NGjVbBgQXXq1EmSFBQUpO7du+v1119XSEiIgoODNWTIEFWpUkWNGzf24TMDAAC+4tPkZufOnWrbtq3t57i4OMXFxemRRx7RpEmT9Nxzz+nixYsaNGiQkpOTVbNmTc2fP1+FChWy3ebtt99WQECAevXqpUuXLqlhw4aaPHmyAgICfPGUAACAj/nNPjfwvuxaCOZP6EPPpO6/IKtVknQ6Odnrj0W9E/yJWeqSsus10G9rbgAAADKC5AYAAJgKyQ0AADAVv93nBgDcYZYah1tl13oHf0If5kwkN17ij8WMte4QO+56iD70TFr954/vl/SYMXkCzIplKQAAYCokNwAAwFRIbgAAgKlQc+Ml/rgeb4ZCuuxYmwFz8sVY9OeaL3+85gEpmLkBAACmQnIDAABMheQGAACYCjU38Gu+XtfPaN0StUIwu6wa476+BiB7YuYGAACYCskNAAAwFZIbAABgKtTcwO/4U72KP+8zAnNxVltihr2qAF9g5gYAAJgKyQ0AADAVkhsAAGAq1NzA7/jTvhbUPHgmdf8FySrJv15jAOZDcoM0+VNhr69QUOyZtPrPzGOLxA3wPZalAACAqZDcAAAAUyG5AQAApkLNDdJE7QBfnAn3efO198eaL64LyA6YuQEAAKZCcgMAAEyF5AYAAJgKNTcwJW/VPfhjzQPM7daaFjaRBDKGmRsAAGAqJDcAAMBUSG4AAICpUHOTQ7EPC+Bb7BcDZB5mbgAAgKmQ3AAAAFMhuQEAAKZCzU0OxXq/a9hnxDOp+y9IVkmMPwCZi+QGyEI5rZA7rU0Qc1o/ZFR220SSpBX+gmUpAABgKiQ3AADAVEhuAACAqVBzA2Qhf6lJoOYFmcGTceUv7w2YAzM3AADAVEhuAACAqZDcAAAAU6HmBsiBsqq+Ias38aOWKPvKrNcuu+0V5G/S6z9/rZVi5gYAAJgKyQ0AADAVkhsAAGAq1NzAtLyxhs96vWf4binf8td6iKzEl996Jrv2HzM3AADAVEhuAACAqZDcAAAAU6HmBqbljXqD7Lre7C+yep8bs2H8ARlDcoM0UfRJQbGnKCi+iWQOyFosSwEAAFMhuQEAAKZCcgMAAEyFmhukyax1Ajmt3gPuM+vYB3IKZm4AAICpkNwAAABTIbkBAACmQs0Nchx36imy0yZq1BJ5j7/0Jfssec6TPqT2Kvti5gYAAJgKyQ0AADAVkhsAAGAq1NyYWK07arNe7yFqHjIHtQyuyU41X/6KPsyZmLkBAACmQnIDAABMheQGAACYCjU3Jvbfc9tZa/YQ6/WeSd1/QbL6LhgAOQbJDdLlLxua+QIFxZ5Jq/9y8phyh7vjj0Jt4CaWpQAAgKn4dXJz/fp1vfnmm7rnnnsUGhqqe+65R2+++aauXbtma2MYhuLi4lSpUiWFhYWpdevW+uWXX3wYNQAA8CW/Tm7GjRunqVOnatSoUdq2bZtGjhypjz76SO+9956tzfjx4xUfH69Ro0Zp9erVCgkJUfv27XX27FkfRg4AAHzFr2tutm3bppYtW6pVq1aSpNKlS6tVq1basWOHpJuzNpMmTdLzzz+vdu3aSZImTZqkyMhIzZ07V7169fJZ7GaSk9fxnRUUUy8Cf5UZYzMnv/+Rffn1zE29evW0YcMG/f7775KkX3/9VevXr9cDDzwgSTpw4ICSkpLUtGlT220CAwMVFRWlrVu3+iRmAADgW349c/P888/r3Llzqlu3rgICAnTt2jUNHDhQvXv3liQlJSVJkkJCQuxuFxISoiNH0v6IQWJiYuYF7Wdy0nPNLKn7sNYdPgoE8AEzXEPM8Bx8yR/7L70tOvw6uZk/f75mzZqlqVOnqlKlSvrxxx/1yiuvKDw8XI8//ritncVisbudYRgOx26VU/YtYY8WzzntQz4ajhwku19DuA56Jrv2n18nN6+//rqeeeYZdezYUZJUpUoVHTx4UGPHjtXjjz+u0NBQSdLRo0dVsmRJ2+2OHz/uMJsDeAP1NvAWV2pZsusvFsDX/Lrm5sKFCwoICLA7FhAQoBs3bki6WWAcGhqqNWvW2M5funRJmzdvVt26dbM0VgAA4B/8euamZcuWGjdunEqXLq1KlSopISFB8fHxiomJkXRzOapfv34aM2aMIiMjFRERodGjR6tgwYLq1KmTj6MHAAC+4NfJzTvvvKO33npLL730ko4fP67Q0FD16NFDL7/8sq3Nc889p4sXL2rQoEFKTk5WzZo1NX/+fBUqVMiHkQMAAF+xJCcnG74OApmD9XrPZaQPqcu5jcr/+y+biJuWv+2Lw3XQM9m1//y65gYAAMBdJDcAAMBUSG4AAICp+HVBMZAd+VvNgS+lXq8PklUSfeSq7FrvAPgayQ0yJKcUzda6Q+xI7IG0+i87jB8SMCD7YlkKAACYCskNAAAwFZIbAABgKtTcIEO8WY+QHeovkPN4e1xSwwNkHWZuAACAqZDcAAAAUyG5AQAApkLNDXzOn2sRfLmJGrVI5pKR15N9ltznz9cTZB1mbgAAgKmQ3AAAAFMhuQEAAKZCzQ1Mz5PaFWoekBlcrQvhizOBjGHmBgAAmArJDQAAMBWSGwAAYCrU3MD03N33gv1lkNlcGWPs1wJkHDM3AADAVEhuAACAqZDcAAAAU6HmBkjl1loH9hnxTOr+C5JVEvUkADIXyU0ORdGsa9jEzzNp9d+t449EB4C3sSwFAABMheQGAACYCskNAAAwFWpucijqHFxDQbG9zKjVov4rbdR8ee52fch10LyYuQEAAKZCcgMAAEyF5AYAAJgKNTfA/zir/aDmAb6QUgtCzZfn6MOciZkbAABgKiQ3AADAVEhuAACAqViSk5MNXweBzMFas+foQ884fHGm1SpJOp2c7PXHYr8cwD/5Yj8hZm4AAICpkNwAAABTIbkBAACmwj43AEzBjN8TRM2X5+hDz2TX/iO5AXK4zCzETWsTRIp/XcMmkp7zpA/NmDDnFCxLAQAAUyG5AQAApkJyAwAATIWaGyCLUW+CzEKNiKPsWhALzzBzAwAATIXkBgAAmArJDQAAMBVqboAs5kldBPU6uB3Gh6Nb97mhJinnYOYGAACYCskNAAAwFZIbAABgKtTcIFvL7BoDvtsHWe3WuhD2aPEcfZgzMXMDAABMheQGAACYCskNAAAwFWpukK1l9r4VrNd7JnX/Bckqif1GAGQukhtkSE7ZLIyCYs+k1X85ZfykhyQPyBwsSwEAAFMhuQEAAKZCcgMAAEyFmhtkSGbXClCTgZwgrXFOLQ7gGWZuAACAqZDcAAAAUyG5AQAApkLNDVxGHQyQNVLea7XukE4r2aexANkRMzcAAMBUSG4AAICpkNwAAABToeYGLsuJe2/4wxdnUuuUs+W01z8nXmfgfczcAAAAUyG5AQAApuL3yc0///yjp59+WuXLl1doaKjq1q2rDRs22M4bhqG4uDhVqlRJYWFhat26tX755RcfRgwAAHzJr2tukpOT1aJFC9WrV0+zZ89WkSJFdODAAYWEhNjajB8/XvHx8YqPj1dkZKTeeecdtW/fXtu3b1ehQoV8GD3gHdm5BiF1zVKQrJKy93PKSv5Q8wVkR36d3EyYMEFhYWGaMmWK7ViZMmVs/28YhiZNmqTnn39e7dq1kyRNmjRJkZGRmjt3rnr16pXVIQNZyt+LTWvdIemI43F/j9tfpNV/7iCRRE7k18tSS5YsUc2aNdWrVy9FRESoQYMG+vDDD2UYhiTpwIEDSkpKUtOmTW23CQwMVFRUlLZu3eqrsAEAgA9laOYmOTlZ8+bN0/79+3Xq1ClbspHCYrHo/fff9zi4/fv3a9q0aerfv7+ef/55/fjjj4qNjZUk9e3bV0lJSZJkt0yV8vORI2n/uZOYmOhxbNlFTnqumcWf+7DWHb6OAP7On8dvVqEPPOOP/Zfecq3byc26devUvXt3nT17VoUKFZLVanVoY7FY3L1bp27cuKF7771XQ4cOlSRVq1ZN+/bt09SpU9W3b980H88wjNvGkFPWsFmv95zf96GHSxYwP78ev1nA79/Dfi679p/byc2rr76q4OBgLVmyRFWrVs2MmGxCQ0NVsWJFu2MVKlTQ33//bTsvSUePHlXJkiVtbY4fP+4wm4OsZ4a6Cm/UPCB782XNSnb9xQL4mts1N3/88Yf69euX6YmNJNWrV09//PGHw+OXKlVKklS6dGmFhoZqzZo1tvOXLl3S5s2bVbdu3UyPDwAA+B+3k5vSpUvr0qVLmRGLg/79+2v79u0aPXq09u3bp6+//loffvihevfuLenmclS/fv00btw4LVy4UD///LP69++vggULqlOnTlkSIwAA8C9uJzcvvviiPvnkE506dSoz4rFTo0YNzZgxQwsWLNB9992nESNG6NVXX7UlN5L03HPPqX///ho0aJCaNGmif/75R/Pnz2ePGwAAcihLcnKykX6z/zd+/Hh9/fXX2r9/v9q3b6+SJUsqICDA/k4tFj377LNeDRTu8/f1ejPU5MBNlf/3XzYRz3I5db8bf78O+rvs2n9uFxQPGzbM9v+ffPKJ0zYkNwAAwFfcTm52796dGXEAAAB4hdvJTXh4eGbEAQAA4BUZ/m6ps2fPasOGDfrrr78k3Ux6GjRoQCEvXJaZNQDU88DXvDG+s2u9A+BrGUpupkyZojfffFPnz5+3++qFggUL6rXXXtNTTz3ltQABAADc4XZyM2vWLL3yyiuqWbOm+vXrp4oVK8owDP3++++aPHmyBg8erODgYHXp0iUz4gUAALgtt5Ob+Ph41a1bV4sXL1bu3P9/86pVq6pdu3Zq06aNJk6cSHIDAAB8wu3kJjExUcOHD7dLbGx3lju3OnToYPuiS8BXvFXPQ82DZ1L3X5CsknLunisAsobbyU3BggWVlJSU5vmkpCQVKFDAo6AAmKMoOq0vHjXDc8sK3vziVhJK5CRuf/1C06ZNNWXKFK1fv97h3IYNG/Thhx8qOjraK8EBAAC4y+2Zm6FDh2rTpk1q166d7rnnHlWoUEGS9PvvvyshIUHFihVjWQoAAPiM2zM3JUuW1Pr169W/f39duHBBCxcu1MKFC3XhwgUNGDBA69evV4kSJTIjVgAAgHS5/cWZyD58WQxLTQWc4oszJble/0JBu+foQ89k1/5ze+YGAADAn6VbczNq1ChZLBYNHDhQuXLl0qhRo9K9U4vFopdfftkrAQIAALgj3WWp4OBgWSwW/fPPP8qbN6+Cg4PTv1OLRSdPnvRakMgYlqXgd1iWksSyVFaiDz2TXfsv3ZmbU6dO3fZnIAUJDZA+9psBMh81NwAAwFTcTm7uvPNOzZkzJ83z8+fP15133ulRUAAAABnldnJjGLf/5PiNGzdksVgyHBAAAIAn3N6hWNJtk5f//ve/slqtGY0H2ZgZawmyazFdRlE3lfnc6WNvfrdUTuXtPjTjdc6MXEpuJk2apMmTJ9t+Hjx4sEaMGOHQ7vTp0zpz5oy6du3qvQgBAADc4FJyU6RIEdtfr3/99ZdCQ0MVFhZm18ZisahAgQK699571bdvX+9HCgAA4AKXkpsuXbqoS5cukqQ2bdpo0KBBatSoUaYGBgAAkBFu19wsXrw4M+IA4Ae8XU+QumYpSNZMeRyzymk1X5mBPsyZMlRQLElXr15VYmKiTp8+rRs3bjicr1+/vkeBAZLvC1wp6PRMWv3n69c1u0jpP5JBwD1uJzeGYeitt97SlClTdP78+TTb8fULAADAF9ze52bChAkaM2aM2rdvr0mTJskwDA0bNkxjx45V5cqVVbVqVS1YsCAzYgUAAEiX28nN559/rjZt2mjChAl64IEHJEnVqlVTjx49tHr1al2/fl0bNmzweqAAAACucHtZ6uDBg+rfv78kKVeum7nRlStXJEn58uVT165dNWXKFA0ZMsSLYSKnurXWgDoN5FRmGfvUDiGruD1zY7VadeHCBUlS4cKFlTdvXh06dMh2Pl++fNTbAAAAn3E7ualcubISEhJu3jhXLtWoUUNTp07VoUOHdPDgQX366ad87A4AAPiM28lN586dlZiYqEuXLkmSXn/9de3du1dVq1ZVtWrVtHfvXr3++uteDxQAAMAVbtfcdOvWTd26dbP9fN9992nz5s1atmyZAgICFB0drfLly3s1SGQts6zvAxnlL7UhbEAHZEyGN/G7VZkyZdSvXz9v3BUAAIBH3F6W2rJli8aOHZvm+bFjx2rbtm0eBQUAAJBRbs/cjBo1SlarNc3zP/30kzZs2KB58+Z5EhcAAECGuJ3cJCQkaODAgWmer127tkaPHu1RUPAtf6k38AdmqXmgjso9/tJfOeG7zbjeIDO4vSx14cIFWSyW27Y5d+5chgMCAADwhNvJTUREhFauXJnm+RUrVqhcuXIeBQUAAJBRbic3jz/+uFavXq0XX3xRJ06csB0/ceKEXnrpJa1du1bdu3f3apAAAACucrvmpk+fPvrxxx/1ySef6NNPP1VISIgsFouOHj0qwzD06KOP8rFwwM/4qq4hdc1SkKw+jSe7MUvNF5DVMrTPzYQJE9S5c2ctXLhQ+/fvl2EYKlu2rNq1a6cGDRp4O0YoYwWOOaEYMbPRh+4haQHgDzK8id/999+v+++/35uxAAAAeMztmhsAAAB/lu7MTZs2bZQrVy7Nnz9fuXPnVtu2bdO9U4vFooULF3olQAAAAHekm9wYhqEbN27Yfr5x40a6+9wYhuF5ZLCTkVoGihE9l9V96C+bx2VU6vjTqlnK7s8zq5ip5ot6LGSldJObJUuW3PZnAAAAf0LNDQAAMBWSGwAAYCrpLksFBwenW2OTmsVisdu9GDAbakYA97jynqEuB96SbnLz8ssvu53cAAAA+Eq6yc3gwYOzIg4AAACvoOYGAACYSoa+fiE5OVnvv/++VqxYoYMHD0qSSpUqpebNm2vAgAEKDg72apBAVnF1nxbAn1G7gpzO7ZmbP/74Q1FRURozZoyuXbumBg0aqH79+rp27ZrGjBmjqKgoJSYmZkasAAAA6XJ75mbQoEE6d+6cvvnmGzVs2NDu3Lp169S9e3fFxsZq/vz5XgsSAADAVW7P3GzdulVPP/20Q2IjSY0aNdJTTz2lLVu2eCU4AAAAd7k9cxMUFCSr1ZrmeavVetvzgLew1wzgnD+/N6gHQlZwe+ame/fu+uKLL3T27FmHc6dPn9YXX3yh7t27eyU4AAAAd7k9cxMZGSmLxaJatWrpkUceUbly5SRJe/fu1axZsxQSEqLIyEgtWLDA7nbt27f3TsQAAAC3YUlOTjbcuYErH/O2WCwyDMPu55MnT7ofHTySmJioyMhIX4eRafx56h1pqPy///7i0yjgQ1m9LGX262Bmy6795/bMzaJFizIjDsBtWXGRzK5vbH+Ruv+CZJVE3YWrGH9Axrid3DRo0CAz4kAmYoYj49jEzzNp9V9OHZMkdUDWyNAOxZJ08eJF7dy5U8eOHVP9+vX1r3/9y5txAQAAZEiGvltq8uTJqlixotq0aaNevXppz549kqQTJ04oPDxc06dP92qQAAAArnI7uZkxY4YGDx6sZs2aaeLEiXaFw0WKFFGTJk0cPikFAACQVdxeloqPj1eLFi308ccfO/0EVPXq1fXRRx95JTh4B+v8GefNgs6cWmeC/+fuGEirZon3NHB7bs/c7N27Vy1atEjzfJEiRXTixAmPggIAAMgot5ObQoUK6fTp02me37t3L8XFAADAZ9xObho2bKgZM2bo8uXLDucOHTqkzz77TM2aNfNKcAAAAO5yu+bmP//5j6Kjo9W4cWM9/PDDslgsWrlypdasWaNPP/1UuXPn1ssvv5wZsQIAlPn1W9T0ILtze+amXLlyWr58ucLCwjRq1CgZhqH4+HiNHz9e1apV04oVKxQWFpYZsQIAAKQrQ/vcVKxYUQsWLNC+ffu0atUqrVy5UomJiZo9e7ZWrVqle++919txSpLGjBkjq9WqQYMG2Y4ZhqG4uDhVqlRJYWFhat26tX75hS+uAQAgp3J5Wery5ctavny5/vzzTwUHB6tly5YKDQ1VjRo1dOHCBU2ZMkWTJk3SsWPHbN8U7k3bt2/XZ599pipVqtgdHz9+vOLj4xUfH6/IyEi98847at++vbZv365ChQp5PQ4AAODfXEpujhw5otatW2v//v22TfsKFiyoWbNmKVeuXOrTp48OHz6sOnXqaMyYMWrTpo1Xgzx9+rT69OmjiRMn6p133rEdNwxDkyZN0vPPP6927dpJkiZNmqTIyEjNnTtXvXr18moc8G+ZUYfAd0shJ8quezJRK4QULi1LjRgxQn/99ZeeffZZffXVV3rnnXcUGBio5557TjExMSpVqpSWLVumb7/9Vm3btpXFYvFqkCnJS6NGjeyOHzhwQElJSWratKntWGBgoKKiorR161avxgAAALIHl2Zu1q5dq27dumnYsGG2Y0WLFlXPnj3VvHlzzZw5U7lyZah8J12fffaZ9u3bpylTpjicS0pKkiSFhITYHQ8JCdGRI2n/uZ2YmOjdIP1YTnqute7wdQQAfCmt611Oug5mBn/sv/R2jncpuTl69Khq1apld6x27dqSpMceeyzTEpvExEQNHz5cy5YtU968edNsl3qmyDCM284eeWs7fX/nza8OyBZYPgJyNGfXuxx3HfSy7Np/LiU3169fV/78+e2OpfxcuHBh70f1P9u2bdOJEyd033332cWyadMmffzxx9qyZYukm8lXyZIlbW2OHz/uMJsD88uM9fbs+sb2F6n7L0hWSdRGuIrxB2SMy5+W2r9/v3bs2GH7+cyZM5JuvvnuuMNxPaBmzZoeB9e6dWuHj5UPGDBA5cuX14svvqiIiAiFhoZqzZo1qlGjhiTp0qVL2rx5s4YPH+7x4yP7yKwCSAqKPZNW/2XXgtWs5q3xRzKJnMbl5CYuLk5xcXEOx1PvRpyyJOTsG8PdZbVaZbVa7Y4VKFBAwcHBuuuuuyRJ/fr105gxYxQZGamIiAiNHj1aBQsWVKdOnTx+fAAAkP24lNzEx8dndhwZ9txzz+nixYsaNGiQkpOTVbNmTc2fP589bgAAyKEsycnJhq+DQObISev1LHNkE5X/9182Ec9SOXlZKiddBzNDdu0/t784E/BHrly8SYCQ2bydRGTXXyyAr2XOZ7gBAAB8hOQGAACYCskNAAAwFWpukGNkpB4iO9Q8UEvkP7z9WrDPkjds93UA8AFmbgAAgKmQ3AAAAFMhuQEAAKZCzQ1Myxv1D9Q8ICulrgvLDjVffi8x0dcRwAeYuQEAAKZCcgMAAEyF5AYAAJgKNTcwLW98zw81D55J3X9BskrK2V/kCCDzkdwAXsamev8vrYJs+sg1WVXQTrIJs2FZCgAAmArJDQAAMBWSGwAAYCrU3ABe5s36BWpTkBX8dZxRC4SMYuYGAACYCskNAAAwFZIbAABgKtTcAJnMX+sZkLNRzwIzY+YGAACYCskNAAAwFZIbAABgKtTcAJksq2sbqPGBK1LGCbU3MCNmbgAAgKmQ3AAAAFMhuQEAAKZCzQ1gMv5UQ5GYmKjIyEjbz0GySvKvGP1Z6v4D4BqSG8CLKOa1V+sOSUccj9NPrkmr/8yO5BeeYlkKAACYCskNAAAwFZIbAABgKtTcINvKirqNnFrzAN9KqTmhoBjIGGZuAACAqZDcAAAAUyG5AQAApkLNDbyCfUsAz7C3C+A9zNwAAABTIbkBAACmQnIDAABMhZobeEV2qBegLgj+zNn4TG+fpezwvgN8gZkbAABgKiQ3AADAVEhuAACAqVBzgxwjI/UJfLePZ1L3X5CskqgVcRXjD8gYkhv4JX8p/uWLMz2TVv/5y+ubmUjgAN9hWQoAAJgKyQ0AADAVkhsAAGAq1NzAL/lLvUJOK+jMCbUwWcUbfZkZNV/+8t4CMhMzNwAAwFRIbgAAgKmQ3AAAAFOh5gaAzeliydTdmJwrry91OcjumLkBAACmQnIDAABMheQGAACYCjU3MBVv14vw3VLIKG/UreS0fZYAb2HmBgAAmArJDQAAMBWSGwAAYCrU3MBUvL0/BzUPnkndf0GySmIfFQCZi+QG2VJWbTRHQbFn0uo/Ngq8PZI/wDMsSwEAAFMhuQEAAKZCcgMAAEyFmhtkS5lVk0AtCPxByjj0Rc0X9T4wA2ZuAACAqZDcAAAAUyG5AQAApkLNDbIM9Swwo8ysUWETSSBjmLkBAACm4tfJzXvvvacmTZqoVKlSKl++vLp27aqff/7Zro1hGIqLi1OlSpUUFham1q1b65dffvFRxAAAwNf8OrnZsGGDnnzySX377bdauHChcufOrYcfflinTp2ytRk/frzi4+M1atQorV69WiEhIWrfvr3Onj3rw8gBAICvWJKTkw1fB+Gqc+fOKTw8XDNmzFCrVq1kGIYqVaqkPn36aODAgZKkixcvKjIyUiNGjFCvXr18HLFvsV7vOX/uw2xZw1T5f/9lctV0/HV/HH9+D2cH2bX//HrmJrVz587pxo0bslqtkqQDBw4oKSlJTZs2tbUJDAxUVFSUtm7d6qMoAQCAL2WrT0u98sorqlq1qurUqSNJSkpKkiSFhITYtQsJCdGRI2lv65mYmJh5QfqZnPRcM4u/9mGtO3wdAfD//PV9Ivl3bNmBP/ZferNJ2Sa5efXVV7VlyxYtX75cAQEBducsFovdz4ZhOBy7VXacYsuI7Dqd6E/8ug+zeFt+4Hb89X3i1+/hbCC79l+2SG4GDx6s+fPna9GiRSpTpozteGhoqCTp6NGjKlmypO348ePHHWZzALPx1xqHW6W+MAbJKil7xO4PsusvFsDX/D65iY2N1fz587V48WJVqFDB7lzp0qUVGhqqNWvWqEaNGpKkS5cuafPmzRo+fLgvwoWH/K1I1hdfXGgmafWfv73O/srd8UfSCNzk18nNwIED9dVXX+mLL76Q1Wq11dgULFhQd9xxhywWi/r166cxY8YoMjJSERERGj16tAoWLKhOnTr5OHoAAOALfp3cTJ06VZLUrl07u+OxsbEaPHiwJOm5557TxYsXNWjQICUnJ6tmzZqaP3++ChUqlOXxAgAA3/Pr5CY5OTndNhaLRYMHD7YlOwAAIGfz6+QGrnNWw0C9CHIKs9aaUFAMZEy22sQPAAAgPSQ3AADAVEhuAACAqVBzYxLOag5Yr/ecO33I3i2+Y9a+p27Oc/7Sh2atC/NXzNwAAABTIbkBAACmQnIDAABMhZobwEv8aU3drDUoQHpSvw+pPcyZmLkBAACmQnIDAABMheQGAACYCjU3yBZ8VUPiL3tkwD9ldp0V9SJAxjBzAwAATIXkBgAAmArJDQAAMBVqbpAt+GoPGWoePJO6/4JkleRfewIBMB+SG8BE/G3zvrQKsv0tTn+VkwraSXjhTSxLAQAAUyG5AQAApkJyAwAATIWaG8ALqCEBXEd9DTIbMzcAAMBUSG4AAICpkNwAAABToeYG8AC1NoDrqLVBVmHmBgAAmArJDQAAMBWSGwAAYCrU3AAeMFsNATVEyEy+GF/Ovp/LbO9bOGLmBgAAmArJDQAAMBWSGwAAYCrU3ACw8XYtQmJioiIjI20/B8maKY9jVqn7D+6jD3MmkhtkCrMUpjorRoTr0uo/s4yPzELyB3iGZSkAAGAqJDcAAMBUSG4AAICpUHMDr6CGAgDgL5i5AQAApkJyAwAATIXkBgAAmAo1N/AKs+7LQS0RfCFl3GXlPktmfQ8jZ2LmBgAAmArJDQAAMBWSGwAAYCrU3AC38d9z2/nSPQ/c+qWF1C/5N39+fagHgruYuQEAAKZCcgMAAEyF5AYAAJgKNTcAssTpYskKktX2/0jfrTVLAFxHcgO4wZ+LLv1RWpvQ0Y+uycpN/NJCIorsiGUpAABgKiQ3AADAVEhuAACAqVBz4wcyq/7AH9brszv6EM5kVR0KBcVAxjBzAwAATIXkBgAAmArJDQAAMBVqbvxAZq3f+2q9nj1M4GvszQLkbMzcAAAAUyG5AQAApkJyAwAATIWaG3idmeodsqpuiTol7zJLf/pqnyUzvYeRMzFzAwAATIXkBgAAmArJDQAAMBVqbgA/YNYah9Q1S0GySjLv8/U2vlsKyBiSG2R7mVk8yhdneiat/jNLwa9Eogb4I5alAACAqZDcAAAAUyG5AQAApkLNDSSZqwYC8BbqaYDsiZkbAABgKqZJbqZOnap77rlHoaGhatSokTZt2uTrkAAAgA+YIrmZP3++XnnlFb300kv6/vvvVadOHXXu3FkHDx70dWgAACCLmaLmJj4+Xo8++qh69OghSXr33Xe1atUqffzxxxo6dKiPo8seqC1wjk3UPMMmfgB8IdvP3Fy5ckW7du1S06ZN7Y43bdpUW7du9VFUAADAV7L9zM2JEyd0/fp1hYSE2B0PCQnR0aNHnd4mMTExK0LzCznpuWYW+tAzdv23fXvKQd8Ekw0x/jxHH3rGH/svvRn1bJ/cpLBYLHY/G4bhcCxFTllmYEnFc/ShZ+g/z9B/nqMPPZNd+y/bL0sVKVJEAQEBDrM0x48fd5jNAQAA5pftk5u8efOqevXqWrNmjd3xNWvWqG7duj6KCgAA+IoplqUGDBigp556SjVr1lTdunX18ccf659//lGvXr18HRoAAMhipkhuOnTooJMnT+rdd99VUlKSKleurNmzZys8PNzXoQEAgCxmiuRGknr37q3evXv7OgwAAOBj2b7mBgAA4FYkNwAAwFRIbgAAgKmQ3AAAAFMhuQEAAKZCcgMAAEyF5AYAAJgKyQ0AADAVkhsAAGAqJDcAAMBUSG4AAICpkNwAAABTIbkBAACmYklOTjZ8HQQAAIC3MHMDAABMheQGAACYCskNAAAwFZIbAABgKiQ3AADAVEhufGTjxo2KiYlR5cqVZbVaNWPGDLvzhmEoLi5OlSpVUlhYmFq3bq1ffvnFrs3ly5c1aNAglStXTsWLF1dMTIwOHTpk1yY5OVl9+/ZVeHi4wsPD1bdvXyUnJ9u1OXjwoLp27arixYurXLlyevnll3XlyhW7Nnv27NGDDz6osLAwVa5cWaNGjZJh+PaDdt7ow9atW8tqtdr9e+KJJ+zamLUP0+u/hQsXqkOHDipfvrysVqvWr1/vcB85eQx6o/8Yf8777+rVqxo6dKiioqJUvHhxVaxYUb1799bBgwft7iMnjz/JO31o1jFIcuMj58+f11133aWRI0cqMDDQ4fz48eMVHx+vUaNGafXq1QoJCVH79u119uxZW5vBgwdr0aJFmjZtmpYuXaqzZ8+qa9euun79uq1N7969lZCQoDlz5mju3LlKSEjQU089ZTt//fp1de3aVefOndPSpUs1bdo0LVy4UEOGDLG1OXPmjNq3b6+iRYtq9erVGjlypCZOnKj3338/k3rHNd7oQ0nq1q2bfvvtN9u/sWPH2p03ax+m138XLlxQnTp19NZbb6V5Hzl5DHqj/yTGn7P+u3Dhgnbv3q2BAwdq3bp1+vLLL3Xo0CF16tRJ165ds7XLyeNP8k4fSuYcg+xz4wdKlCihd955R926dZN0c8ahUqVK6tOnjwYOHChJunjxoiIjIzVixAj16tVLp0+fVkREhOLj49WlSxdJ0t9//62qVatq7ty5io6O1m+//aa6detq+fLlqlevniRp8+bNatWqlbZv367IyEitXLlSXbp00Y8//qiSJUtKkr766is9++yzSkxMVOHChTVt2jQNGzZMv//+u+0N9O677+rjjz/Wzz//LIvFktVd5iAjfSjd/Kvlrrvu0rvvvuv0fnNKH6buv1udOHFC5cuX16JFi3T//ffbjjMG/19G+k9i/KW4Xf+l+PXXX1WvXj1t3LhRVapUYfylkpE+lMw7Bpm58UMHDhxQUlKSmjZtajsWGBioqKgobd26VZK0a9cuXb161a5NyZIlVbFiRVubbdu26Y477lDdunVtberVq6eCBQvatalYsaJtQEpSdHS0Ll++rF27dtna3HfffXZ/GURHR+vIkSM6cOCA9zvAC1zpwxTz5s1TuXLlVK9ePf3nP/+xm9nJyX2YHsagdzD+XJPSL1arVRLjLyNS92EKM47B3F6/R3gsKSlJkhQSEmJ3PCQkREeOHJEkHT16VAEBASpSpIhDm6NHj9raFClSxC4jtlgs+te//mXXJvXjFClSRAEBAXZtihcv7vA4KefKlCnjydPNFK70oSR17txZpUqVUlhYmH799Ve98cYb+umnn/T1119Lytl9mB7GoOcYf665cuWK/vOf/6hly5YqUaKEJMafu5z1oWTeMUhy48dST9MZhpHu1F3qNs7au9Im9XFnsdzutv4ivT7s2bOn7f+rVKmiMmXKKDo6Wrt27VL16tWd3oez+zFzH7qLMeg6xl/6rl27pr59++r06dOaOXNmuu0Zf45u14dmHYMsS/mh0NBQSbJlvCmOHz9uy3SLFi2q69ev68SJE7dtc/z4cbtqdMMwdOLECbs2qR/nxIkTun79+m3bHD9+XJLjzIi/cKUPnbn33nsVEBCgffv2ScrZfZgexqD3Mf7sXbt2TU8++aT27Nmjb775RnfeeaftHOPPNbfrQ2fMMgZJbvxQ6dKlFRoaqjVr1tiOXbp0SZs3b7ate1avXl158uSxa3Po0CFb8Zck1alTR+fOndO2bdtsbbZt26bz58/btfntt9/sPj65Zs0a5cuXz5a116lTR5s3b9alS5fs2hQrVkylS5f2fgd4gSt96MyePXt0/fp1W3KUk/swPYxB72P8/b+rV6+qV69e2rNnjxYtWmTrkxSMv/Sl14fOmGUMktz4yLlz55SQkKCEhATduHFDf//9txISEnTw4EFZLBb169dP48aN08KFC/Xzzz+rf//+KliwoDp16iRJCgoKUvfu3fX6669r7dq12r17t5566ilVqVJFjRs3liRVrFhRzZo10wsvvKDt27dr27ZteuGFF9SiRQtFRkZKkpo2barKlSvr6aef1u7du7V27Vq9/vrrevzxx1W4cGFJUqdOnRQYGKj+/fvr559/1sKFCzVu3Dj179/fp1Oynvbhn3/+qVGjRmnnzp06cOCAVqxYoSeffFL33HOP7VMBZu7D2/WfJJ06dUoJCQn6+eefJd3sr4SEBFs9U04fg572H+Mv7f67du2aevToof/+97+aOnWqLBaLkpKSlJSUpIsXL0pi/Eme96Gpx2BycrLBv6z/t2jRIkOSw79HHnnESE5ONk6dOmXExsYaoaGhRr58+YyoqChj06ZNdvfxzz//GH369DGCg4ONwMBAo0WLFsZPP/1k1+bPP/80unTpYhQqVMgoVKiQ0aVLF2P//v12bX788UejRYsWRmBgoBEcHGz06dPHSEpKsmuzceNG47777jPy5ctnhIaGGq+88opx6tSpbN2HP/30kxEVFWUEBwcbefPmNcqWLWs89dRTxp9//pkj+jC9/ouPj3d6PjY2ljHohf5j/KXdf7t373Z6TpIRHx/P+PNSH5p5DLLPDQAAMBWWpQAAgKmQ3AAAAFMhuQEAAKZCcgMAAEyF5AYAAJgKyQ0AADAVkhsAHouLi3P4puGqVauqX79+vglIN7eIb9iwoUaMGJFu2/Xr18tqtWr9+vVuP86MGTNktVrtvtm4X79+qlq1qtv35akrV66ocuXKmjZtWpY/NuBPSG4Ak0j5JWu1WvX99987bdO0aVNZrVbVrl07i6PLel9//bX++OMPnyZYWS1v3rzq37+/Ro8ebduFFsiJSG4Ak8mfP7/mzJnjcHzv3r364YcflD9//iyJ47///a8mTJiQJY/lzIQJE/TQQw/pX//6l89i8IXu3bvr+PHjTscAkFOQ3AAm07x5c33zzTe6fPmy3fGvvvpKRYsW1b333pslceTLl0958uTJksdKbc+ePdq5c6c6duzok8f3tgsXLrjc1mq1qkmTJvryyy8zMSLAv5HcACbTsWNHnTt3TsuXL7c7PnfuXHXo0EG5cjl/28+bN0/R0dEqVqyYwsPD1bVrV/36668O7b799lvVr19foaGhqlmzpqZPn+70/lLX3Fy5ckVvvfWWGjdurNKlSyssLEzR0dFaunSpw22tVqteeOEFrVy5Uvfff79CQ0NVo0YNzZ0716U+WLx4sXLnzq0GDRo4nEtMTFSHDh1UrFgxVahQQa+99pquXr3q9H6WLl1q65PSpUurW7du+v33312KIbUZM2aoXbt2qlChgooWLaqaNWtq3LhxunHjhl271q1bq3bt2vrpp5/Utm1bFS9eXC+99JIkad++ferZs6cqVqyo0NBQValSRT169NDhw4ft7qNRo0basmWLjh07lqFYgewut68DAOBdxYsXV/369TVnzhy1a9dO0s0lon379qlLly768ccfHW4zbtw4DRs2TG3btlVMTIzOnz+vqVOnqkWLFlq3bp3KlCkjSVq3bp0effRRlStXTkOGDNGlS5c0YsQIhYaGphvX2bNn9cknn6h9+/Z67LHHdPHiRc2ZM0fdunXT3LlzFR0dbdd++/btWrJkiXr16qXu3btr+vTp6tu3r6pWraqKFSve9rG2bNmiihUrKjAw0O74sWPH1Lp1a128eFEDBgzQnXfeqVmzZmnNmjUO9zF37lz16dNHd999t4YMGaIzZ87oww8/VPPmzbV27Vpbn7jqo48+UmRkpJo1a6bAwECtWbNGw4YN05kzZ/T666/btT19+rQ6dOigtm3bqmPHjgoKCtLVq1fVoUMHXbp0Sb1791ZoaKiSkpK0evVqHT58WMWLF7fdvkaNGrZ+aNu2rVtxAmZAcgOYUOfOnTVw4EAlJyfLarXqq6++Uvny5W2/9G518OBBvfnmm4qNjdXgwYNtx2NiYlSnTh2NHj1a77//viTp9ddfl9Vq1YoVKxQcHCxJateunaKiotKNyWq1as+ePcqXL5/t2FNPPaX7779fEydOdEhufv31V23cuNGWyDz88MO6++679cUXX6T7CajExETdc889DsfHjRuno0ePaunSpbaYe/bs6TDDc/XqVQ0ZMkQRERFavny5ChYsKOnmrEqTJk309ttv68MPP0z3Od9q6dKlKlCggO3n3r1769///remTJmi2NhYu345evSoRo4cqaefftp27Mcff9T+/fv12Wef2ZJWSRo0aJDDY6UkXhmdZQKyO5alABNq166dLBaLvvnmG127dk1ff/21Onfu7LTtokWLdO3aNXXs2FEnTpyw/cuTJ49q1apl++RVUlKSdu/erZiYGFtiI0kVK1Z0SEycCQgIsP0Cv3Llik6dOqWzZ8+qfv362rVrl0P7+++/326GpmjRooqMjNT+/fvTfawTJ044fDRdurmkVq1aNbtkrECBAurZs6ddu127dikpKUlPPvmkLbGRpGrVqqlx48ZasWKFDMNIN45bpSQ2169fV3Jysk6cOKEGDRro/PnzSkxMtGubO3duh5gKFSokSVq1apXOnz9/28dKeX1OnDjhVoyAWTBzA5hQUFCQmjdvrtmzZ6t48eI6duxYmsnN3r17JUl16tRxej7ll/Jff/0lSYqMjHRoExERoRUrVqQb1/Tp0/XBBx/ot99+s0sOLBaLQ9tSpUo5HLNarTp16lS6jyPJafJx8OBBp8s0ERERdj+nPNcKFSo4tK1YsaJWr16tM2fOKCgoyKVYJGnz5s0aPny4duzYoStXrtidO336tN3PYWFhDp9qK1OmjJ5++mlNnjxZs2fPVt26ddWiRQt17dpVRYoUsWub8tyd9SuQE5DcACbVuXNn9ejRQ5JUs2ZNlS9f3mm7lILWuXPnKndux0tCSgHy7X5hujKLMXfuXD377LNq1aqVnnvuOYWEhCh37tyaMWOG048tBwQEOL0fVx6rSJEiSk5Odnouo/FnpG2K/fv3q3379ipXrpzi4uJUsmRJ5cuXT7t379bQoUMdiopT1wqlGDlypHr06KFly5Zp9erVeu211zR69GgtWbJElStXtrVLee6pkx4gpyC5AUyqRYsWKly4sDZu3KiRI0em2a5s2bKSpJIlS6pSpUpptitdurQk53UcKbM/tzN//nyVKVNGX375pV2CMWPGjHRv664KFSrY7RicolSpUi7FHx4eLunmc23atKnducTERFmtVhUuXNjleJYuXapLly5p1qxZtvuW5DTG9FSuXFmVK1fWiy++qJ9++kmNGzfWpEmT7PYU+vPPPyU5n3kCcgJqbgCTypcvn8aMGaPY2Fh16tQpzXYPPfSQcufOrbi4OIcZBEk6fvy4JCk0NFT33HOPZs2aZbc09Ntvv2nVqlXpxpMyE3PrzMf+/fu1ePFil5+Tq+rWravffvvNYX+Y5s2ba/fu3dq0aZPt2IULF/Tpp5/atatevbpCQ0P18ccf2+30++OPP2rNmjVq3ry5W0s+zp775cuX3SpKPnPmjK5du2Z3LOUTYalnqX744QdJN/sByImYuQFM7HZJTYoyZcrojTfe0JAhQ9SsWTO1bdtWwcHBOnjwoFasWKFatWpp7NixkqQ33nhDHTt2VPPmzfX444/r4sWL+uijj1S5cmX99NNPt32cVq1aadGiRXrkkUfUqlUrHT58WNOmTVP58uXTva27Wrdurbi4OK1fv14tWrSwHX/++ec1e/ZsxcTEqG/fvraPgt/6KSZJypMnj9566y316dPHVteS8lHwwoUL69VXX3UrnujoaOXNm1cxMTHq2bOnrly5olmzZqW555Az33//vQYNGqSHHnpIkZGRMgxD8+fP19mzZx02K1y7dq3q1KmjkJAQt+IEzILkBoAGDBigiIgITZw4Ue+9956uXbumYsWKqV69eurevbutXZMmTTRjxgyNGDFCI0aMUKlSpfTaa6/p0KFD6SYojz76qI4fP65p06Zp7dq1KleunN5++23t27fP68nN3XffrerVq2vevHl2yU3RokW1ZMkSxcbG6v3331ehQoXUtWtXNW3aVB06dLC7j06dOikwMFBjxozRiBEjlDdvXjVo0EDDhg1ze4+biIgIzZgxQ8OHD9fQoUNVpEgRxcTEqEGDBmrfvr3Lz6lZs2ZauXKlpk+frnz58qly5cqaMWOGWrdubWt36tQprVu3TqNHj3YrRsBMLMnJye5XxwGAn5s/f74GDBighISEHDWDMXHiRMXHx2vnzp1pFiYDZkfNDQBTat++vSIjIzVp0iRfh5Jlrly5og8++EADBw4ksUGOxswNAAAwFWZuAACAqZDcAAAAUyG5AQAApkJyAwAATIXkBgAAmArJDQAAMBWSGwAAYCokNwAAwFT+D9mOh01n9LNNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "replication_number = np.arange(1, 101)\n",
    "\n",
    "replication_number = replication_number.astype(str)\n",
    "\n",
    "intervals2 = pd.DataFrame(np.array([left_ends, right_ends]), columns=[replication_number])\n",
    "\n",
    "intervals2\n",
    "\n",
    "plt.figure(figsize=(8,8))\n",
    "for i in np.arange(100):\n",
    "    ends = intervals2.iloc[:,i]\n",
    "    plt.plot(ends, np.array([i+1, i+1]), color='gold')\n",
    "plt.plot(np.array([pop_median, pop_median]), np.array([0, 100]), color='red', lw=2)\n",
    "plt.xlabel('Median (dollars)')\n",
    "plt.ylabel('Replication')\n",
    "plt.title('Population Median and Intervals of Estimates');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To summarize what the simulation shows, suppose you are estimating the population median by the following process: \n",
    "\n",
    "- Draw a large random sample from the population.\n",
    "- Bootstrap your random sample and get an estimate from the new random sample. \n",
    "- Repeat the above step thousands of times, and get thousands of estimates.\n",
    "- Pick off the \"middle 95%\" interval of all the estimates.\n",
    "\n",
    "That gives you one interval of estimates. Now if you repeat **the entire process** 100 times, ending up with 100 intervals, then about 95 of those 100 intervals will contain the population parameter.\n",
    "\n",
    "In other words, this process of estimation captures the parameter about 95% of the time. \n",
    "\n",
    "You can replace 95% by a different value, as long as it's not 100. Suppose you replace 95% by 80% and keep the sample size fixed at 500. Then your intervals of estimates will be shorter than those we simulated here, because the \"middle 80%\" is a smaller range than the \"middle 95%\". Only about 80% of your intervals will contain the parameter."
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}