census_analysis_2019_11_30.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CENSUS - Resort Town Tax Incentive Investigation\n",
    "  \n",
    "  Kalin Gibbons  \n",
    "  Carlos Segura Cerna  \n",
    "  Mark Baehr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "## Required Python Modules\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import math\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "from census import Census\n",
    "from us import states # Latest python=3.6.x\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "\n",
    "import locale\n",
    "locale.setlocale( locale.LC_ALL, '' );"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Has to be separate from the import cell to work\n",
    "InteractiveShell.ast_node_interactivity = \"all\"\n",
    "pd.set_option('display.float_format', lambda x: '%.3f' % x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "## Data import\n",
    "---\n",
    "\n",
    "Current data sources include the 2010 census data, census-derived spending data from the George Wathington University Policy Institure, and the Local Option Tax revenue data provided by a subset of the 14 Idaho resort towns. The census data is accessed using a python module that wraps the government's census API inside simpler python method calls. Some useful links concerning available census variables, and the python modules used to access them, are provided in the following links:\n",
    "\n",
    "* See [Initial Data Description](https://docs.google.com/document/d/1P0KDC9JeTSeFnT9454elcvP6jx3g0Svd7wPPidcbivE/edit?usp=sharing) for links to the census documentation.\n",
    "* See [Python Census package](https://pypi.org/project/census/)\n",
    "* See [Python us package](https://pypi.org/project/us/)\n",
    "* See [census_definitions.pdf](https://drive.google.com/a/boisestate.edu/file/d/1CzdsmBPiqQmgiJW6DmjLTWG_P8dpcpfD/view?usp=sharing) for more detailed descriptions of the census variables.\n",
    "\n",
    "These python packages interact with the census APIs using FIPs codes and have them built-in down to the county level, but we have to get the town-level FIPs codes from an excel file published by the government."
   ]
  },
  {
   "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>Summary Level</th>\n",
       "      <th>State Code (FIPS)</th>\n",
       "      <th>County Code (FIPS)</th>\n",
       "      <th>County Subdivision Code (FIPS)</th>\n",
       "      <th>Place Code (FIPS)</th>\n",
       "      <th>Consolidtated City Code (FIPS)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area Name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Hailey city</th>\n",
       "      <td>162</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34390</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ketchum city</th>\n",
       "      <td>162</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>43030</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Summary Level  State Code (FIPS)  County Code (FIPS)  \\\n",
       "Area Name                                                            \n",
       "Hailey city             162                 16                   0   \n",
       "Ketchum city            162                 16                   0   \n",
       "\n",
       "              County Subdivision Code (FIPS)  Place Code (FIPS)  \\\n",
       "Area Name                                                         \n",
       "Hailey city                                0              34390   \n",
       "Ketchum city                               0              43030   \n",
       "\n",
       "              Consolidtated City Code (FIPS)  \n",
       "Area Name                                     \n",
       "Hailey city                                0  \n",
       "Ketchum city                               0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Area Name\n",
       "Hailey city     34390\n",
       "Ketchum city    43030\n",
       "Name: Place Code (FIPS), dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Point python to the data folder, the census year of interest, and the list of \n",
    "# pilot towns.\n",
    "datDir = os.path.join(os.getcwd(), 'data')\n",
    "yearOfInterest = 2010\n",
    "popLim = 10e3\n",
    "townList = ('Hailey', 'Ketchum')\n",
    "townList = [town + ' city' for town in townList]\n",
    "\n",
    "# FIPS and PLACE codes\n",
    "# -----------------------------------------------------------------------------\n",
    "# State codes are easily called from the state module.\n",
    "fipsIdaho = states.ID.fips\n",
    "\n",
    "# Town codes must be imported.\n",
    "geoCodes = pd.read_excel(\n",
    "    os.path.join(datDir,'all_geocodes_v2016.xlsx'), header=4)\n",
    "geoCodes.rename(\n",
    "    columns={'Area Name (including legal/statistical area description)':'Area Name'}, \n",
    "    inplace=True)\n",
    "geoCodes.set_index('Area Name', inplace=True)\n",
    "townCodes = geoCodes.loc[townList, :]\n",
    "\n",
    "# Make sure the towns are in Idaho.\n",
    "idahoMask = townCodes.loc[:,'State Code (FIPS)'] == int(fipsIdaho)\n",
    "townFips = townCodes.loc[idahoMask, 'Place Code (FIPS)']\n",
    "\n",
    "# ---\n",
    "townCodes\n",
    "townFips"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "The cities shown within the filtered dataframe are all from Idaho. Now we can use their FIPS codes to pull data from the Census API. For a first pass, we'll use the population by housing unit, which should include populations excluding hotels, vacation rentals, etc. We'll start with the populations at the state- and county- levels, then handle the town revenues and populations.\n",
    "\n",
    "### State and county populations"
   ]
  },
  {
   "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>name</th>\n",
       "      <th>housingUnit</th>\n",
       "      <th>popByHousingUnit</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Idaho</td>\n",
       "      <td>667796.000</td>\n",
       "      <td>1538631.000</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    name  housingUnit  popByHousingUnit state\n",
       "0  Idaho   667796.000       1538631.000    16"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>name</th>\n",
       "      <th>housingUnit</th>\n",
       "      <th>popByHousingUnit</th>\n",
       "      <th>state</th>\n",
       "      <th>county</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Bear Lake County, Idaho</td>\n",
       "      <td>3914.000</td>\n",
       "      <td>5955.000</td>\n",
       "      <td>16</td>\n",
       "      <td>007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Benewah County, Idaho</td>\n",
       "      <td>4629.000</td>\n",
       "      <td>9215.000</td>\n",
       "      <td>16</td>\n",
       "      <td>009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Ada County, Idaho</td>\n",
       "      <td>159471.000</td>\n",
       "      <td>382651.000</td>\n",
       "      <td>16</td>\n",
       "      <td>001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Adams County, Idaho</td>\n",
       "      <td>2636.000</td>\n",
       "      <td>3955.000</td>\n",
       "      <td>16</td>\n",
       "      <td>003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Bannock County, Idaho</td>\n",
       "      <td>33191.000</td>\n",
       "      <td>81047.000</td>\n",
       "      <td>16</td>\n",
       "      <td>005</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      name  housingUnit  popByHousingUnit state county\n",
       "0  Bear Lake County, Idaho     3914.000          5955.000    16    007\n",
       "1    Benewah County, Idaho     4629.000          9215.000    16    009\n",
       "2        Ada County, Idaho   159471.000        382651.000    16    001\n",
       "3      Adams County, Idaho     2636.000          3955.000    16    003\n",
       "4    Bannock County, Idaho    33191.000         81047.000    16    005"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 2010 census data\n",
    "# ------------------------------------------------------------------------------\n",
    "# * acs5: ACS 5 Year Estimates (2016, 2015, 2014, 2013, 2012, 2011, 2010)\n",
    "# * acs1dp: ACS 1 Year Estimates, Data Profiles (2016, 2015, 2014, 2013, 2012)\n",
    "# * sf1: Census Summary File 1 (2010, 2000, 1990)\n",
    "# * sf3: Census Summary File 3 (2000, 1990)\n",
    "\n",
    "census = Census('2ac5e0436e9326797fc8e7129ff3a00c85a7278a')\n",
    "sf1 = census.sf1\n",
    "\n",
    "# Add desirable variables to the census_vars.xlsx spreadsheet, as needed.\n",
    "censusCipher = pd.read_excel(os.path.join(datDir, 'census_vars.xlsx'))\n",
    "censusCipher = censusCipher.loc[:, ['api_var', 'dev_var']].set_index('api_var')\n",
    "censusCipher = censusCipher.to_dict()['dev_var']\n",
    "\n",
    "# Import the state-level data\n",
    "idahoCensusData = pd.DataFrame(sf1.state(list(censusCipher.keys()),\n",
    "                                         fipsIdaho, year=yearOfInterest))\n",
    "idahoCensusData.rename(columns=censusCipher, inplace=True)\n",
    "\n",
    "# Import the county-level data\n",
    "countyCensusData = pd.DataFrame(sf1.state_county(list(censusCipher.keys()), \n",
    "                                                fipsIdaho, \n",
    "                                                Census.ALL, \n",
    "                                                year=yearOfInterest))\n",
    "countyCensusData.rename(columns=censusCipher, inplace=True)\n",
    "\n",
    "# ---\n",
    "idahoCensusData.head()\n",
    "countyCensusData.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "The data imported from the Census API looks correct. Next we can import the local option tax revenues as well as the census-derived revenue for the state of Idaho.\n",
    "\n",
    "### Local Option Tax Revenue"
   ]
  },
  {
   "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>2010</th>\n",
       "      <th>2011</th>\n",
       "      <th>2012</th>\n",
       "      <th>2013</th>\n",
       "      <th>2014</th>\n",
       "      <th>2015</th>\n",
       "      <th>2016</th>\n",
       "      <th>2017</th>\n",
       "      <th>2018</th>\n",
       "      <th>2019</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Town</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Hailey</th>\n",
       "      <td>311253.240</td>\n",
       "      <td>315603.980</td>\n",
       "      <td>335014.160</td>\n",
       "      <td>348850.260</td>\n",
       "      <td>422566.620</td>\n",
       "      <td>475799.930</td>\n",
       "      <td>543967.990</td>\n",
       "      <td>585178.010</td>\n",
       "      <td>593277.430</td>\n",
       "      <td>630565.680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ketchum</th>\n",
       "      <td>1451069.880</td>\n",
       "      <td>1638904.070</td>\n",
       "      <td>1755694.500</td>\n",
       "      <td>1774377.860</td>\n",
       "      <td>2961925.740</td>\n",
       "      <td>3807206.470</td>\n",
       "      <td>4027047.670</td>\n",
       "      <td>4370168.720</td>\n",
       "      <td>4646897.970</td>\n",
       "      <td>4729557.790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stanley</th>\n",
       "      <td>87716</td>\n",
       "      <td>128103</td>\n",
       "      <td>142159</td>\n",
       "      <td>147903</td>\n",
       "      <td>160590</td>\n",
       "      <td>202542</td>\n",
       "      <td>224716</td>\n",
       "      <td>247277</td>\n",
       "      <td>273394.000</td>\n",
       "      <td>259108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lava Hot Springs</th>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>28635.170</td>\n",
       "      <td>30251.750</td>\n",
       "      <td>36481.400</td>\n",
       "      <td>44575.970</td>\n",
       "      <td>47788.440</td>\n",
       "      <td>50859.340</td>\n",
       "      <td>48688.380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Driggs</th>\n",
       "      <td>192330</td>\n",
       "      <td>180472</td>\n",
       "      <td>193370</td>\n",
       "      <td>217175</td>\n",
       "      <td>235563</td>\n",
       "      <td>247975</td>\n",
       "      <td>275360</td>\n",
       "      <td>305360</td>\n",
       "      <td>371954.000</td>\n",
       "      <td>514895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Victor</th>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>---</td>\n",
       "      <td>108730</td>\n",
       "      <td>127225</td>\n",
       "      <td>135066</td>\n",
       "      <td>417217.000</td>\n",
       "      <td>450000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Riggins</th>\n",
       "      <td>62299</td>\n",
       "      <td>58052</td>\n",
       "      <td>58248</td>\n",
       "      <td>59453</td>\n",
       "      <td>57445</td>\n",
       "      <td>67644</td>\n",
       "      <td>57731</td>\n",
       "      <td>94858</td>\n",
       "      <td>119236.000</td>\n",
       "      <td>117442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ponderay</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>295000.000</td>\n",
       "      <td>---</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        2010        2011        2012        2013        2014  \\\n",
       "Town                                                                           \n",
       "Hailey            311253.240  315603.980  335014.160  348850.260  422566.620   \n",
       "Ketchum          1451069.880 1638904.070 1755694.500 1774377.860 2961925.740   \n",
       "Stanley                87716      128103      142159      147903      160590   \n",
       "Lava Hot Springs         ---         ---         ---   28635.170   30251.750   \n",
       "Driggs                192330      180472      193370      217175      235563   \n",
       "Victor                   ---         ---         ---         ---         ---   \n",
       "Riggins                62299       58052       58248       59453       57445   \n",
       "Ponderay                 ---         ---         ---         ---         ---   \n",
       "\n",
       "                        2015        2016        2017        2018        2019  \n",
       "Town                                                                          \n",
       "Hailey            475799.930  543967.990  585178.010  593277.430  630565.680  \n",
       "Ketchum          3807206.470 4027047.670 4370168.720 4646897.970 4729557.790  \n",
       "Stanley               202542      224716      247277  273394.000      259108  \n",
       "Lava Hot Springs   36481.400   44575.970   47788.440   50859.340   48688.380  \n",
       "Driggs                247975      275360      305360  371954.000      514895  \n",
       "Victor                108730      127225      135066  417217.000      450000  \n",
       "Riggins                67644       57731       94858  119236.000      117442  \n",
       "Ponderay                 ---         ---         ---  295000.000         ---  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>Program</th>\n",
       "      <th>Amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1275769000.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>332836933.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>255423930.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>197553689.390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>299697316.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program         Amount\n",
       "0             Medical Assistance Program (Medicaid) 1275769000.000\n",
       "1                      Federal Direct Student Loans  332836933.000\n",
       "2         Supplemental Nutrition Assistance Program  255423930.000\n",
       "3  Medicare Supplemental Medical Insurance (Part B)  197553689.390\n",
       "4                 Highway Planning and Construction  299697316.000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1 b) Local Option Tax revenue from resort towns, 2010-2019\n",
    "# ------------------------------------------------------------------------------\n",
    "lotTax = pd.read_excel(os.path.join(datDir, 'towns_lot_revenues.xlsx'))\n",
    "lotTax.set_index('Town', inplace=True)\n",
    "\n",
    "\n",
    "# 1 e) Census-derived revenue from the State of Idaho\n",
    "# ------------------------------------------------------------------------------\n",
    "idahoCensusFunds = pd.read_excel(\n",
    "    os.path.join(datDir, 'gwu_by_state_fy2016_dev_friendly.xlsx'), header=2)\n",
    "idahoCensusFunds = idahoCensusFunds.loc[:, ['Program','Idaho']]\n",
    "idahoCensusFunds.rename(columns={'Idaho':'Amount'}, inplace=True)\n",
    "\n",
    "# ---\n",
    "lotTax\n",
    "idahoCensusFunds.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "No issues importing this data, but any analysis will require the census-derived funding at the town-level. To date, none of the contacted towns have reported receiving direct funding from any of the programs specified within the GWU dataset. Most programs distribute funds through various state or county committees / offices.\n",
    "\n",
    "### Census-derived funding estimation\n",
    "\n",
    "As a first pass, we can estimate town funding using a simple population proportion. We'll assume whatever proportion of Idaho's population lives within a given town, that is the proportion of funding they will receive. \n",
    "\n",
    "#### Town-level estimates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>name</th>\n",
       "      <th>housingUnit</th>\n",
       "      <th>popByHousingUnit</th>\n",
       "      <th>state</th>\n",
       "      <th>place</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Hailey city, Idaho</td>\n",
       "      <td>3527.000</td>\n",
       "      <td>7894.000</td>\n",
       "      <td>16</td>\n",
       "      <td>34390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ketchum city, Idaho</td>\n",
       "      <td>3564.000</td>\n",
       "      <td>2689.000</td>\n",
       "      <td>16</td>\n",
       "      <td>43030</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  name  housingUnit  popByHousingUnit state  place\n",
       "0   Hailey city, Idaho     3527.000          7894.000    16  34390\n",
       "1  Ketchum city, Idaho     3564.000          2689.000    16  43030"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "array([0.00513053, 0.00174766])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>Program</th>\n",
       "      <th>Total</th>\n",
       "      <th>Hailey city, Idaho</th>\n",
       "      <th>Ketchum city, Idaho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1275769000.000</td>\n",
       "      <td>6545377.343</td>\n",
       "      <td>2229607.255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>332836933.000</td>\n",
       "      <td>1707631.491</td>\n",
       "      <td>581684.961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>255423930.000</td>\n",
       "      <td>1310461.380</td>\n",
       "      <td>446393.546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>197553689.390</td>\n",
       "      <td>1013556.092</td>\n",
       "      <td>345256.186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>299697316.000</td>\n",
       "      <td>1537607.531</td>\n",
       "      <td>523768.261</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program          Total  \\\n",
       "0             Medical Assistance Program (Medicaid) 1275769000.000   \n",
       "1                      Federal Direct Student Loans  332836933.000   \n",
       "2         Supplemental Nutrition Assistance Program  255423930.000   \n",
       "3  Medicare Supplemental Medical Insurance (Part B)  197553689.390   \n",
       "4                 Highway Planning and Construction  299697316.000   \n",
       "\n",
       "   Hailey city, Idaho  Ketchum city, Idaho  \n",
       "0         6545377.343          2229607.255  \n",
       "1         1707631.491           581684.961  \n",
       "2         1310461.380           446393.546  \n",
       "3         1013556.092           345256.186  \n",
       "4         1537607.531           523768.261  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Census-derived revenue from resort towns, 2010-2019\n",
    "# ------------------------------------------------------------------------------\n",
    "\n",
    "# Import the GWU state-level census spending data for Idaho (only 2016 data)\n",
    "# Done, dataframe=idahoCensusFunds[\"Program\",\"Amount\"]\n",
    "\n",
    "# Calculate town populations from the year 2010\n",
    "townCensusData = pd.DataFrame()\n",
    "for place, town in zip(townFips.tolist(), townFips.index):\n",
    "    dummyDf = pd.DataFrame(sf1.state_place(list(censusCipher.keys()), \n",
    "                                           fipsIdaho, \n",
    "                                           place, \n",
    "                                           year=yearOfInterest))\n",
    "    townCensusData = pd.concat([townCensusData, dummyDf])\n",
    "\n",
    "del(dummyDf)\n",
    "townCensusData.rename(columns=censusCipher, inplace=True)\n",
    "townCensusData.reset_index(drop=True, inplace=True)\n",
    "\n",
    "# Distribute state-level spending to towns by proportion of state population.\n",
    "populationProportion = (townCensusData.loc[:, 'popByHousingUnit'].to_numpy() \n",
    "                        / idahoCensusData.loc[:, 'popByHousingUnit'].to_numpy())\n",
    "\n",
    "townsCensusRevenue2016 = idahoCensusFunds.copy().rename(columns={'Amount':'Total'})\n",
    "for idx in townCensusData.index:\n",
    "    iName = townCensusData.loc[idx, 'name']\n",
    "    townsCensusRevenue2016[iName] = (townsCensusRevenue2016.loc[:, 'Total'].to_numpy() \n",
    "                               * populationProportion[idx])\n",
    "        \n",
    "    \n",
    "# ---\n",
    "townCensusData\n",
    "populationProportion\n",
    "townsCensusRevenue2016.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "The GWU dataset was for the year 2016, only. So, we now have estimated town-level funding for one year, but we need to adjust this value for all the other years in the 2010 decade. We'll start by assuming that the funding amounts kept pace with an average inflationof 3% per year."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year 2016\n"
     ]
    },
    {
     "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>Program</th>\n",
       "      <th>Total</th>\n",
       "      <th>Hailey city, Idaho</th>\n",
       "      <th>Ketchum city, Idaho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1275769000.000</td>\n",
       "      <td>6545377.343</td>\n",
       "      <td>2229607.255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>332836933.000</td>\n",
       "      <td>1707631.491</td>\n",
       "      <td>581684.961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>255423930.000</td>\n",
       "      <td>1310461.380</td>\n",
       "      <td>446393.546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>197553689.390</td>\n",
       "      <td>1013556.092</td>\n",
       "      <td>345256.186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>299697316.000</td>\n",
       "      <td>1537607.531</td>\n",
       "      <td>523768.261</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program          Total  \\\n",
       "0             Medical Assistance Program (Medicaid) 1275769000.000   \n",
       "1                      Federal Direct Student Loans  332836933.000   \n",
       "2         Supplemental Nutrition Assistance Program  255423930.000   \n",
       "3  Medicare Supplemental Medical Insurance (Part B)  197553689.390   \n",
       "4                 Highway Planning and Construction  299697316.000   \n",
       "\n",
       "   Hailey city, Idaho  Ketchum city, Idaho  \n",
       "0         6545377.343          2229607.255  \n",
       "1         1707631.491           581684.961  \n",
       "2         1310461.380           446393.546  \n",
       "3         1013556.092           345256.186  \n",
       "4         1537607.531           523768.261  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year 2017\n"
     ]
    },
    {
     "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>Program</th>\n",
       "      <th>Total</th>\n",
       "      <th>Hailey city, Idaho</th>\n",
       "      <th>Ketchum city, Idaho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1314042070.000</td>\n",
       "      <td>6741738.663</td>\n",
       "      <td>2296495.473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>342822040.990</td>\n",
       "      <td>1758860.436</td>\n",
       "      <td>599135.510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>263086647.900</td>\n",
       "      <td>1349775.221</td>\n",
       "      <td>459785.352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>203480300.071</td>\n",
       "      <td>1043962.775</td>\n",
       "      <td>355613.872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>308688235.480</td>\n",
       "      <td>1583735.757</td>\n",
       "      <td>539481.309</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program          Total  \\\n",
       "0             Medical Assistance Program (Medicaid) 1314042070.000   \n",
       "1                      Federal Direct Student Loans  342822040.990   \n",
       "2         Supplemental Nutrition Assistance Program  263086647.900   \n",
       "3  Medicare Supplemental Medical Insurance (Part B)  203480300.071   \n",
       "4                 Highway Planning and Construction  308688235.480   \n",
       "\n",
       "   Hailey city, Idaho  Ketchum city, Idaho  \n",
       "0         6741738.663          2296495.473  \n",
       "1         1758860.436           599135.510  \n",
       "2         1349775.221           459785.352  \n",
       "3         1043962.775           355613.872  \n",
       "4         1583735.757           539481.309  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def simulate_census_revenue_for_locations(locationCensusRevenue, scalar, columns):\n",
    "    '''Multiply a column of funding data by a scalar value.'''\n",
    "    df = locationCensusRevenue.copy()\n",
    "    for col in columns:\n",
    "        df[col] = df[col].apply(lambda x: x*scalar)\n",
    "    return df\n",
    "\n",
    "# Let's calculate the Census-derived revenue from resort towns, 2010-2019\n",
    "#townCensusRevenue2016\n",
    "columns = list(townsCensusRevenue2016.filter(regex=('city|Total')).columns)\n",
    "\n",
    "# Inflation increases amounts since 2016\n",
    "townsCensusRevenue2017 = simulate_census_revenue_for_locations(townsCensusRevenue2016, 1.03, columns)\n",
    "townsCensusRevenue2018 = simulate_census_revenue_for_locations(townsCensusRevenue2017, 1.03, columns)\n",
    "townsCensusRevenue2019 = simulate_census_revenue_for_locations(townsCensusRevenue2018, 1.03, columns)\n",
    "\n",
    "# Inflation decreases amounts before 2016\n",
    "townsCensusRevenue2015 = simulate_census_revenue_for_locations(townsCensusRevenue2016, 0.97, columns)\n",
    "townsCensusRevenue2014 = simulate_census_revenue_for_locations(townsCensusRevenue2015, 0.97, columns)\n",
    "townsCensusRevenue2013 = simulate_census_revenue_for_locations(townsCensusRevenue2014, 0.97, columns)\n",
    "townsCensusRevenue2012 = simulate_census_revenue_for_locations(townsCensusRevenue2013, 0.97, columns)\n",
    "townsCensusRevenue2011 = simulate_census_revenue_for_locations(townsCensusRevenue2012, 0.97, columns)\n",
    "townsCensusRevenue2010 = simulate_census_revenue_for_locations(townsCensusRevenue2011, 0.97, columns)\n",
    "\n",
    "# ---\n",
    "print('Year 2016')\n",
    "townsCensusRevenue2016.head()\n",
    "print('Year 2017')\n",
    "townsCensusRevenue2017.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "That's a 3% increase, so we can continue on to the county-level data. \n",
    "\n",
    "### County-level estimates\n",
    "\n",
    "This data exists for five programs, but there are another 49 programs without data; we'll use the same estimation techniques for each program."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Use the five programs with data to quantify error associated with \n",
    "# estimated revenue."
   ]
  },
  {
   "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>Program</th>\n",
       "      <th>Total</th>\n",
       "      <th>Bannock County, Idaho</th>\n",
       "      <th>Blaine County, Idaho</th>\n",
       "      <th>Bonner County, Idaho</th>\n",
       "      <th>Idaho County, Idaho</th>\n",
       "      <th>Custer County, Idaho</th>\n",
       "      <th>Teton County, Idaho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1275769000.000</td>\n",
       "      <td>67200810.424</td>\n",
       "      <td>17510169.587</td>\n",
       "      <td>33600819.791</td>\n",
       "      <td>13053442.552</td>\n",
       "      <td>3604352.079</td>\n",
       "      <td>8426738.020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>332836933.000</td>\n",
       "      <td>17532101.530</td>\n",
       "      <td>4568249.536</td>\n",
       "      <td>8766158.925</td>\n",
       "      <td>3405528.575</td>\n",
       "      <td>940343.817</td>\n",
       "      <td>2198461.977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>255423930.000</td>\n",
       "      <td>13454391.114</td>\n",
       "      <td>3505741.503</td>\n",
       "      <td>6727278.561</td>\n",
       "      <td>2613452.433</td>\n",
       "      <td>721633.597</td>\n",
       "      <td>1687131.873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>197553689.390</td>\n",
       "      <td>10406090.781</td>\n",
       "      <td>2711461.561</td>\n",
       "      <td>5203109.588</td>\n",
       "      <td>2021334.376</td>\n",
       "      <td>558136.348</td>\n",
       "      <td>1304886.061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>299697316.000</td>\n",
       "      <td>15786480.560</td>\n",
       "      <td>4113402.056</td>\n",
       "      <td>7893337.671</td>\n",
       "      <td>3066449.880</td>\n",
       "      <td>846716.485</td>\n",
       "      <td>1979567.435</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program          Total  \\\n",
       "0             Medical Assistance Program (Medicaid) 1275769000.000   \n",
       "1                      Federal Direct Student Loans  332836933.000   \n",
       "2         Supplemental Nutrition Assistance Program  255423930.000   \n",
       "3  Medicare Supplemental Medical Insurance (Part B)  197553689.390   \n",
       "4                 Highway Planning and Construction  299697316.000   \n",
       "\n",
       "   Bannock County, Idaho  Blaine County, Idaho  Bonner County, Idaho  \\\n",
       "0           67200810.424          17510169.587          33600819.791   \n",
       "1           17532101.530           4568249.536           8766158.925   \n",
       "2           13454391.114           3505741.503           6727278.561   \n",
       "3           10406090.781           2711461.561           5203109.588   \n",
       "4           15786480.560           4113402.056           7893337.671   \n",
       "\n",
       "   Idaho County, Idaho  Custer County, Idaho  Teton County, Idaho  \n",
       "0         13053442.552           3604352.079          8426738.020  \n",
       "1          3405528.575            940343.817          2198461.977  \n",
       "2          2613452.433            721633.597          1687131.873  \n",
       "3          2021334.376            558136.348          1304886.061  \n",
       "4          3066449.880            846716.485          1979567.435  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Census-derived revenue from counties containing resort towns, 2010-2019\n",
    "# ------------------------------------------------------------------------------\n",
    "\n",
    "# Select the counties with resort towns which we have information about.\n",
    "countyCensusData_withResortTowns = countyCensusData[countyCensusData['name'].isin([\n",
    "    'Blaine County, Idaho', \n",
    "    'Custer County, Idaho', \n",
    "    'Bannock County, Idaho', \n",
    "    'Teton County, Idaho', \n",
    "    'Idaho County, Idaho', \n",
    "    'Bonner County, Idaho'])].reset_index()\n",
    "\n",
    "# Distribute state-level spending to counties by proportion of state population.\n",
    "populationProportion_counties = (countyCensusData_withResortTowns.loc[:, 'popByHousingUnit'].to_numpy() \n",
    "                        / idahoCensusData.loc[:, 'popByHousingUnit'].to_numpy())\n",
    "\n",
    "# Calculate the census-derived revenue for the counties for the year 2016\n",
    "countiesCensusRevenue2016 = idahoCensusFunds.copy().rename(columns={'Amount':'Total'})\n",
    "for idx in countyCensusData_withResortTowns.index:\n",
    "    iName = countyCensusData_withResortTowns.loc[idx, 'name']\n",
    "    countiesCensusRevenue2016[iName] = (countiesCensusRevenue2016.loc[:, 'Total'].to_numpy() \n",
    "                               * populationProportion_counties[idx])\n",
    "countiesCensusRevenue2016.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year 2016\n"
     ]
    },
    {
     "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>Program</th>\n",
       "      <th>Total</th>\n",
       "      <th>Bannock County, Idaho</th>\n",
       "      <th>Blaine County, Idaho</th>\n",
       "      <th>Bonner County, Idaho</th>\n",
       "      <th>Idaho County, Idaho</th>\n",
       "      <th>Custer County, Idaho</th>\n",
       "      <th>Teton County, Idaho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1275769000.000</td>\n",
       "      <td>67200810.424</td>\n",
       "      <td>17510169.587</td>\n",
       "      <td>33600819.791</td>\n",
       "      <td>13053442.552</td>\n",
       "      <td>3604352.079</td>\n",
       "      <td>8426738.020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>332836933.000</td>\n",
       "      <td>17532101.530</td>\n",
       "      <td>4568249.536</td>\n",
       "      <td>8766158.925</td>\n",
       "      <td>3405528.575</td>\n",
       "      <td>940343.817</td>\n",
       "      <td>2198461.977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>255423930.000</td>\n",
       "      <td>13454391.114</td>\n",
       "      <td>3505741.503</td>\n",
       "      <td>6727278.561</td>\n",
       "      <td>2613452.433</td>\n",
       "      <td>721633.597</td>\n",
       "      <td>1687131.873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>197553689.390</td>\n",
       "      <td>10406090.781</td>\n",
       "      <td>2711461.561</td>\n",
       "      <td>5203109.588</td>\n",
       "      <td>2021334.376</td>\n",
       "      <td>558136.348</td>\n",
       "      <td>1304886.061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>299697316.000</td>\n",
       "      <td>15786480.560</td>\n",
       "      <td>4113402.056</td>\n",
       "      <td>7893337.671</td>\n",
       "      <td>3066449.880</td>\n",
       "      <td>846716.485</td>\n",
       "      <td>1979567.435</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program          Total  \\\n",
       "0             Medical Assistance Program (Medicaid) 1275769000.000   \n",
       "1                      Federal Direct Student Loans  332836933.000   \n",
       "2         Supplemental Nutrition Assistance Program  255423930.000   \n",
       "3  Medicare Supplemental Medical Insurance (Part B)  197553689.390   \n",
       "4                 Highway Planning and Construction  299697316.000   \n",
       "\n",
       "   Bannock County, Idaho  Blaine County, Idaho  Bonner County, Idaho  \\\n",
       "0           67200810.424          17510169.587          33600819.791   \n",
       "1           17532101.530           4568249.536           8766158.925   \n",
       "2           13454391.114           3505741.503           6727278.561   \n",
       "3           10406090.781           2711461.561           5203109.588   \n",
       "4           15786480.560           4113402.056           7893337.671   \n",
       "\n",
       "   Idaho County, Idaho  Custer County, Idaho  Teton County, Idaho  \n",
       "0         13053442.552           3604352.079          8426738.020  \n",
       "1          3405528.575            940343.817          2198461.977  \n",
       "2          2613452.433            721633.597          1687131.873  \n",
       "3          2021334.376            558136.348          1304886.061  \n",
       "4          3066449.880            846716.485          1979567.435  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year 2017\n"
     ]
    },
    {
     "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>Program</th>\n",
       "      <th>Total</th>\n",
       "      <th>Bannock County, Idaho</th>\n",
       "      <th>Blaine County, Idaho</th>\n",
       "      <th>Bonner County, Idaho</th>\n",
       "      <th>Idaho County, Idaho</th>\n",
       "      <th>Custer County, Idaho</th>\n",
       "      <th>Teton County, Idaho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Medical Assistance Program (Medicaid)</td>\n",
       "      <td>1314042070.000</td>\n",
       "      <td>69216834.736</td>\n",
       "      <td>18035474.675</td>\n",
       "      <td>34608844.385</td>\n",
       "      <td>13445045.828</td>\n",
       "      <td>3712482.641</td>\n",
       "      <td>8679540.161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Federal Direct Student Loans</td>\n",
       "      <td>342822040.990</td>\n",
       "      <td>18058064.576</td>\n",
       "      <td>4705297.022</td>\n",
       "      <td>9029143.693</td>\n",
       "      <td>3507694.432</td>\n",
       "      <td>968554.132</td>\n",
       "      <td>2264415.836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Supplemental Nutrition Assistance Program</td>\n",
       "      <td>263086647.900</td>\n",
       "      <td>13858022.848</td>\n",
       "      <td>3610913.748</td>\n",
       "      <td>6929096.918</td>\n",
       "      <td>2691856.006</td>\n",
       "      <td>743282.605</td>\n",
       "      <td>1737745.829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Medicare Supplemental Medical Insurance (Part B)</td>\n",
       "      <td>203480300.071</td>\n",
       "      <td>10718273.504</td>\n",
       "      <td>2792805.407</td>\n",
       "      <td>5359202.876</td>\n",
       "      <td>2081974.407</td>\n",
       "      <td>574880.439</td>\n",
       "      <td>1344032.643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Highway Planning and Construction</td>\n",
       "      <td>308688235.480</td>\n",
       "      <td>16260074.976</td>\n",
       "      <td>4236804.118</td>\n",
       "      <td>8130137.801</td>\n",
       "      <td>3158443.377</td>\n",
       "      <td>872117.980</td>\n",
       "      <td>2038954.458</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Program          Total  \\\n",
       "0             Medical Assistance Program (Medicaid) 1314042070.000   \n",
       "1                      Federal Direct Student Loans  342822040.990   \n",
       "2         Supplemental Nutrition Assistance Program  263086647.900   \n",
       "3  Medicare Supplemental Medical Insurance (Part B)  203480300.071   \n",
       "4                 Highway Planning and Construction  308688235.480   \n",
       "\n",
       "   Bannock County, Idaho  Blaine County, Idaho  Bonner County, Idaho  \\\n",
       "0           69216834.736          18035474.675          34608844.385   \n",
       "1           18058064.576           4705297.022           9029143.693   \n",
       "2           13858022.848           3610913.748           6929096.918   \n",
       "3           10718273.504           2792805.407           5359202.876   \n",
       "4           16260074.976           4236804.118           8130137.801   \n",
       "\n",
       "   Idaho County, Idaho  Custer County, Idaho  Teton County, Idaho  \n",
       "0         13445045.828           3712482.641          8679540.161  \n",
       "1          3507694.432            968554.132          2264415.836  \n",
       "2          2691856.006            743282.605          1737745.829  \n",
       "3          2081974.407            574880.439          1344032.643  \n",
       "4          3158443.377            872117.980          2038954.458  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate the census-derived revenue for counties from year 2010 to 2019\n",
    "# set g: annual rate of change in funding\n",
    "g = 0.03\n",
    "columns = ['Total', 'Blaine County, Idaho', 'Custer County, Idaho', 'Bannock County, Idaho', 'Teton County, Idaho', 'Idaho County, Idaho', 'Bonner County, Idaho']\n",
    "\n",
    "countiesCensusRevenue2017 = simulate_census_revenue_for_locations(countiesCensusRevenue2016, 1+g, columns)\n",
    "countiesCensusRevenue2018 = simulate_census_revenue_for_locations(countiesCensusRevenue2017, 1+g, columns)\n",
    "countiesCensusRevenue2019 = simulate_census_revenue_for_locations(countiesCensusRevenue2018, 1+g, columns)\n",
    "\n",
    "countiesCensusRevenue2015 = simulate_census_revenue_for_locations(countiesCensusRevenue2016, 1-g, columns)\n",
    "countiesCensusRevenue2014 = simulate_census_revenue_for_locations(countiesCensusRevenue2015, 1-g, columns)\n",
    "countiesCensusRevenue2013 = simulate_census_revenue_for_locations(countiesCensusRevenue2014, 1-g, columns)\n",
    "countiesCensusRevenue2012 = simulate_census_revenue_for_locations(countiesCensusRevenue2013, 1-g, columns)\n",
    "countiesCensusRevenue2011 = simulate_census_revenue_for_locations(countiesCensusRevenue2012, 1-g, columns)\n",
    "countiesCensusRevenue2010 = simulate_census_revenue_for_locations(countiesCensusRevenue2011, 1-g, columns)\n",
    "\n",
    "# ---\n",
    "print('Year 2016')\n",
    "countiesCensusRevenue2016.head()\n",
    "print('Year 2017')\n",
    "countiesCensusRevenue2017.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "That's a 3% increase year-over-year. We can now move on to analyzing our simulated data.\n",
    "\n",
    "## Initial investigation\n",
    "\n",
    "---\n",
    "\n",
    "Our initial analysis will focus on estimating the marginal costs and benefits associated with the census / LOT option for a single town: Hailey, Idaho. This town has the most complete data, and is also in danger of exceeding the 10,000 person LOT option population limit. We'll pretend the above data was not simulated, and \n",
    "\n",
    "1. Estimate spending for the 2021-2030 decade\n",
    "\n",
    "    * federal spending for the 2021-2030 decade\n",
    "    \n",
    "    * marginal financial impact of an aditional person being counted on the City of Hailey\n",
    "\n",
    "    * marginal financial impact of an additional person being counted on Blaine County\n",
    "\n",
    "    * marginal financial impact of an additional person being counted on the State of Idaho\n",
    "\n",
    "    * resort tax revenue for 2021-2030\n",
    "\n",
    "1. Construct an integrated model of costs and benefits for the involved entities\n",
    "\n",
    "    * Perform a break-even analysis at which additional census-derived funding offsets lost LOT option revenue\n",
    "    \n",
    "    * Repeat this analysis at the county-level, collecting the towns by county\n",
    "\n",
    "    * Repeat for the state-level, with all towns collected"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Estimated spending for the 2021-2030 decade\n",
    "\n",
    "Start with Hailey City / Blaine County"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "With a growth rate of 1.031, Blaine County should receive $591,647,683 of additional funding within the next decade.\n"
     ]
    }
   ],
   "source": [
    "BlaineGrowth = pow(countiesCensusRevenue2019['Blaine County, Idaho'].sum()\n",
    "                   /countiesCensusRevenue2010['Blaine County, Idaho'].sum()\n",
    "                   ,1/9.0)\n",
    "BlaineFutureFunding = 0\n",
    "for i in range(2,11): # loop over years 2021-2030\n",
    "    BlaineFutureFunding += (countiesCensusRevenue2019['Blaine County, Idaho'].sum()*(pow(BlaineGrowth,i)))\n",
    "\n",
    "print(('With a growth rate of {:.3f}, Blaine County should receive ${:0,.0f}' \n",
    "      + ' of additional funding within the next decade.')\n",
    "      .format(BlaineGrowth, BlaineFutureFunding))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "The growth rate matches the simulated inflation rate, as expected. Next, we can estimate the resort tax revenue for the coming decade."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "With a growth rate of 1.082, Hailey City should receive $9,273,743 of additional funding within the next decade.\n"
     ]
    }
   ],
   "source": [
    "# Estimating LOT revenue for years 2021-2030\n",
    "# Identical effect for the state, assuming all funds received by the state are disbursed\n",
    "HaileyLotGrowth = pow((lotTax.loc['Hailey'][2019]/lotTax.loc['Hailey'][2010]),1/9.0)\n",
    "HaileyFutureFunding = 0\n",
    "for i in range(2,11): # loop over years 2021-2030\n",
    "    HaileyFutureFunding += (lotTax.loc['Hailey'][2019]*pow(HaileyLotGrowth,i))\n",
    "\n",
    "print(('With a growth rate of {:.3f}, Hailey City should receive ${:0,.0f}' \n",
    "      + ' of additional funding within the next decade.')\n",
    "      .format(HaileyLotGrowth, HaileyFutureFunding))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "The LOT revenue data for Hailey was complete, so the growth rate really *did* exceed inflation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hailey's funding after losing the LOT option will break even when population exceeds 10124 people.\n"
     ]
    }
   ],
   "source": [
    "# FIXME: Remove hard-coded magic numbers -> Generalize for loop of every town/county\n",
    "HaileyIdx = 0;\n",
    "BlaineIdx = 6;\n",
    "\n",
    "# calculating per capita funding in Blaine County\n",
    "BlainePerCapFund = BlaineFutureFunding/countyCensusData.iloc[BlaineIdx]['popByHousingUnit']\n",
    "\n",
    "# calculating the approximate percentage of Blaine County's population residing in Hailey\n",
    "HaileyBlainePopRatio = (townCensusData.iloc[HaileyIdx]['popByHousingUnit']\n",
    "                        /countyCensusData.iloc[BlaineIdx]['popByHousingUnit'])\n",
    "\n",
    "# Calculating the number of additional person counted beyond 10000 such that the extra census funding\n",
    "# outweighs the lost LOT revenue as a result of losing resort town status\n",
    "# NOTE: Assumes 100% of census driven revenue is directly proportional to population\n",
    "Threshold = (HaileyFutureFunding/BlainePerCapFund)*HaileyBlainePopRatio\n",
    "\n",
    "print(('Hailey\\'s funding after losing the LOT option will break even when'\n",
    "       + ' population exceeds {:0.0f} people.').format(popLim + Threshold))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "So, if census-derived funding was perfectly proportional to population, and that funding kept exact pace with our estimated inflation, Hailey City would not have financial incentives to ensure accurate census counts until their population exceeded the limit by more that 100 people.\n",
    "\n",
    "Let's perform the same calculation for a range of funding values, and plot the population results. We'll assume that funding proportional to population is a conservative estimate, and reduce that funding \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Finding thresholds with more realisic assumptions regarding funding determination\n",
    "nPoints = 20\n",
    "thresholds = np.zeros(nPoints)\n",
    "revenueFraction = np.zeros(nPoints)\n",
    "for i in range(0, nPoints):\n",
    "    revenueFraction[i] = (i + 1) / float(nPoints)\n",
    "    thresholds[i] = math.ceil(popLim\n",
    "                              + (HaileyFutureFunding / BlainePerCapFund)\n",
    "                              * HaileyBlainePopRatio\n",
    "                              / revenueFraction[i])\n",
    "    \n",
    "# Graphing population thresholds for given fractions of census determined revenues\n",
    "fig = plt.figure(dpi=120)\n",
    "ax = plt.axes(\n",
    "    xlabel = 'Fraction of Estimated Revenue',\n",
    "    ylabel='Population',\n",
    "    title='Break-Even Population for Hailey, ID',\n",
    "    xlim=(0, revenueFraction.max() + 0.05))\n",
    "\n",
    "popLin = plt.scatter(y=thresholds,x=revenueFraction);\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "<br>\n",
    "\n",
    "It looks like the break-even population decays exponentially as census-derived funding increases. For any town performing this analysis, it will be critical to quantify their census-derived funding. We'll end this milestone with a list of those population values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>Population Threshold</th>\n",
       "      <th>Revenue Fraction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12475.000</td>\n",
       "      <td>0.050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11238.000</td>\n",
       "      <td>0.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10825.000</td>\n",
       "      <td>0.150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10619.000</td>\n",
       "      <td>0.200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10495.000</td>\n",
       "      <td>0.250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>10413.000</td>\n",
       "      <td>0.300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>10354.000</td>\n",
       "      <td>0.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>10310.000</td>\n",
       "      <td>0.400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>10275.000</td>\n",
       "      <td>0.450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10248.000</td>\n",
       "      <td>0.500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10225.000</td>\n",
       "      <td>0.550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>10207.000</td>\n",
       "      <td>0.600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>10191.000</td>\n",
       "      <td>0.650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>10177.000</td>\n",
       "      <td>0.700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>10165.000</td>\n",
       "      <td>0.750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>10155.000</td>\n",
       "      <td>0.800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>10146.000</td>\n",
       "      <td>0.850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>10138.000</td>\n",
       "      <td>0.900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>10131.000</td>\n",
       "      <td>0.950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>10124.000</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Population Threshold  Revenue Fraction\n",
       "0              12475.000             0.050\n",
       "1              11238.000             0.100\n",
       "2              10825.000             0.150\n",
       "3              10619.000             0.200\n",
       "4              10495.000             0.250\n",
       "5              10413.000             0.300\n",
       "6              10354.000             0.350\n",
       "7              10310.000             0.400\n",
       "8              10275.000             0.450\n",
       "9              10248.000             0.500\n",
       "10             10225.000             0.550\n",
       "11             10207.000             0.600\n",
       "12             10191.000             0.650\n",
       "13             10177.000             0.700\n",
       "14             10165.000             0.750\n",
       "15             10155.000             0.800\n",
       "16             10146.000             0.850\n",
       "17             10138.000             0.900\n",
       "18             10131.000             0.950\n",
       "19             10124.000             1.000"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#producing the above results in a table\n",
    "HaileyDF = pd.DataFrame(thresholds)\n",
    "HaileyDF['Revenue Fraction'] = revenueFraction\n",
    "HaileyDF.rename( columns = {\n",
    "    0 : 'Population Threshold'\n",
    "}, inplace = True)\n",
    "HaileyDF"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}