d-wasserman/Elasticity_Derivation_Of_Low_Stress_Accessibility_To_Bike_Boardings.ipynb

## Elasticity_Derivation_Of_Low_Stress_Accessibility_To_Bike_Boardings.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Sound Transit Caltrain TCRP 153 Calibration\n",
    "This notebook attempts to augment and replicate the regression analysis methodology documented in TCRP 153 for bicycle ridership by using various measures of bicycling accessiblity using level of traffic stress (Mineta, 2012). This notebook starts with some exploratory charting, tabulations, correlation analysis to help select test variables, and a regression/elasticity analysis to determine potential modification factors. \n",
    "\n",
    "The outputs of this process are elasticities of low stress cycling access to jobs and population to daily, pm, and am bicycle boarding/ridership numbers to Caltrain stations. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Declare functions/ Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import arcpy\n",
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import statsmodels.api as sm\n",
    "import ipywidgets as widgets\n",
    "arcpy.env.overwriteOutput = True\n",
    "sns.set(rc={'figure.figsize':(10,10)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def arcgis_table_to_df(in_fc, input_fields, query=\"\"):\n",
    "    \"\"\"Function will convert an arcgis table into a pandas dataframe with an object ID index, and the selected\n",
    "    input fields using an arcpy.da.SearchCursor.\"\"\"\n",
    "    OIDFieldName = arcpy.Describe(in_fc).OIDFieldName\n",
    "    final_fields = [OIDFieldName] + input_fields\n",
    "    data = [row for row in arcpy.da.SearchCursor(in_fc,final_fields,where_clause=query)]\n",
    "    fc_dataframe = pd.DataFrame(data,columns=final_fields)\n",
    "    fc_dataframe = fc_dataframe.set_index(OIDFieldName,drop=True)\n",
    "    return fc_dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_elasticity(model,regression_df):\n",
    "    \"\"\"Given a dataframe and a model from statsmodels, this function will derive an elasticity for every \n",
    "    column with an index match to the dependent. Elastcities are derived assuming they are derived from some\n",
    "    form of linear regression (elasticity = coeff * (mean(x_col)/mean(y_col)).\n",
    "    @param - model - statsmodel fit result\n",
    "    @param - regression_df - dataframe with source data\"\"\"\n",
    "    dependent_name = str(model.summary().tables[0][0][1])\n",
    "    dependent_mean = regression_df[dependent_name].mean()\n",
    "    columns = [\"Variable Type\", \"Variable\",\"Coefficient\",\"Variable Mean\",\"Significance\",\"Elasticity\"]\n",
    "    data = []\n",
    "    for independent in model.params.index.values:\n",
    "        independent_coeff = model.params.loc[independent]\n",
    "        independent_mean = regression_df[str(independent)].mean()\n",
    "        p_value = model.pvalues.loc[independent]\n",
    "        elasticity = independent_coeff * (independent_mean/dependent_mean)\n",
    "        data.append([\"Independent\",independent,independent_coeff,independent_mean,p_value,elasticity]) \n",
    "    data.append([\"Dependent\",dependent_name,None,dependent_mean,None,None])\n",
    "    elasticity_df = pd.DataFrame(data,columns=columns)\n",
    "    elasticity_df.set_index(\"Variable Type\")\n",
    "    return elasticity_df.style.bar(subset=[\"Elasticity\"], align='mid', color=['#d65f5f', '#5fba7d'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Declare Paths to Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "gdb = r\"N:\\Projects\\Non_SanJose_Projects\\SE Projects\\SE18-0623.00_STSysAccessStratPlan\\Model\\Project_Data.gdb\"\n",
    "fds = os.path.join(gdb,\"BaseData\")\n",
    "stops_data = os.path.join(os.path.join(fds, \"Stops_With_Access_Val_And_BikingData\"))"
   ]
  },
  {
   "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>BART</th>\n",
       "      <th>N_S_ORDER</th>\n",
       "      <th>STATUS</th>\n",
       "      <th>index</th>\n",
       "      <th>FF_Bike_Raw_Min_FF</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_JBS</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_POP</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_HH</th>\n",
       "      <th>FF_Bike_LTS_Walk_FF</th>\n",
       "      <th>...</th>\n",
       "      <th>PS_POP_BLKGRP_ALRND</th>\n",
       "      <th>PS_HH_BLKGRP_ALRND</th>\n",
       "      <th>PS_B01001e1</th>\n",
       "      <th>PS_B11001e1</th>\n",
       "      <th>PS_B08301e1</th>\n",
       "      <th>PS_B08301e18</th>\n",
       "      <th>PS_B08301e19</th>\n",
       "      <th>PS_B25044e1</th>\n",
       "      <th>PS_B25044e10</th>\n",
       "      <th>PS_B25044e3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OBJECTID</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",
       "      <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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4th and King</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>649.273087</td>\n",
       "      <td>153207.748871</td>\n",
       "      <td>78579.951128</td>\n",
       "      <td>38569.412104</td>\n",
       "      <td>497.071654</td>\n",
       "      <td>...</td>\n",
       "      <td>15357.574219</td>\n",
       "      <td>8365.437500</td>\n",
       "      <td>11049.640625</td>\n",
       "      <td>5791.404785</td>\n",
       "      <td>6460.674805</td>\n",
       "      <td>328.934082</td>\n",
       "      <td>1452.118164</td>\n",
       "      <td>5791.404785</td>\n",
       "      <td>1854.352051</td>\n",
       "      <td>280.775940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>22nd St</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>644.589682</td>\n",
       "      <td>78687.501191</td>\n",
       "      <td>67256.143389</td>\n",
       "      <td>29548.603369</td>\n",
       "      <td>539.502826</td>\n",
       "      <td>...</td>\n",
       "      <td>9256.043945</td>\n",
       "      <td>4022.760010</td>\n",
       "      <td>8285.339844</td>\n",
       "      <td>3626.171387</td>\n",
       "      <td>5166.641113</td>\n",
       "      <td>403.354156</td>\n",
       "      <td>410.178406</td>\n",
       "      <td>3626.171387</td>\n",
       "      <td>320.021942</td>\n",
       "      <td>66.299080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Bayshore</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>458.048895</td>\n",
       "      <td>18238.375967</td>\n",
       "      <td>47519.678726</td>\n",
       "      <td>15609.312288</td>\n",
       "      <td>280.229338</td>\n",
       "      <td>...</td>\n",
       "      <td>6736.562500</td>\n",
       "      <td>1881.152710</td>\n",
       "      <td>7119.014648</td>\n",
       "      <td>1994.968506</td>\n",
       "      <td>3319.922363</td>\n",
       "      <td>2.154067</td>\n",
       "      <td>91.777306</td>\n",
       "      <td>1994.968506</td>\n",
       "      <td>213.593430</td>\n",
       "      <td>143.926300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>South San Francisco</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>237.823679</td>\n",
       "      <td>18017.802353</td>\n",
       "      <td>20213.189639</td>\n",
       "      <td>6508.689384</td>\n",
       "      <td>70.799246</td>\n",
       "      <td>...</td>\n",
       "      <td>4752.971680</td>\n",
       "      <td>1443.358765</td>\n",
       "      <td>5364.205078</td>\n",
       "      <td>1576.304443</td>\n",
       "      <td>2900.467041</td>\n",
       "      <td>46.356518</td>\n",
       "      <td>356.708252</td>\n",
       "      <td>1576.304443</td>\n",
       "      <td>254.681091</td>\n",
       "      <td>20.526436</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>San Bruno</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>308.779801</td>\n",
       "      <td>17115.861176</td>\n",
       "      <td>25991.145208</td>\n",
       "      <td>8624.150747</td>\n",
       "      <td>186.223285</td>\n",
       "      <td>...</td>\n",
       "      <td>7981.845703</td>\n",
       "      <td>2237.658447</td>\n",
       "      <td>7120.717285</td>\n",
       "      <td>2132.366699</td>\n",
       "      <td>3871.077148</td>\n",
       "      <td>53.870941</td>\n",
       "      <td>173.980942</td>\n",
       "      <td>2132.366699</td>\n",
       "      <td>96.113113</td>\n",
       "      <td>33.129517</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 64 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         Name  BART  N_S_ORDER  STATUS  index  \\\n",
       "OBJECTID                                                        \n",
       "1                4th and King     0          2       0      0   \n",
       "2                     22nd St     0          4       0      1   \n",
       "3                    Bayshore     0          7       0      2   \n",
       "4         South San Francisco     0          8       0      3   \n",
       "5                   San Bruno     0          9       0      4   \n",
       "\n",
       "          FF_Bike_Raw_Min_FF  ACCESS_Bike_Raw_Min_JBS  \\\n",
       "OBJECTID                                                \n",
       "1                 649.273087            153207.748871   \n",
       "2                 644.589682             78687.501191   \n",
       "3                 458.048895             18238.375967   \n",
       "4                 237.823679             18017.802353   \n",
       "5                 308.779801             17115.861176   \n",
       "\n",
       "          ACCESS_Bike_Raw_Min_POP  ACCESS_Bike_Raw_Min_HH  \\\n",
       "OBJECTID                                                    \n",
       "1                    78579.951128            38569.412104   \n",
       "2                    67256.143389            29548.603369   \n",
       "3                    47519.678726            15609.312288   \n",
       "4                    20213.189639             6508.689384   \n",
       "5                    25991.145208             8624.150747   \n",
       "\n",
       "          FF_Bike_LTS_Walk_FF     ...       PS_POP_BLKGRP_ALRND  \\\n",
       "OBJECTID                          ...                             \n",
       "1                  497.071654     ...              15357.574219   \n",
       "2                  539.502826     ...               9256.043945   \n",
       "3                  280.229338     ...               6736.562500   \n",
       "4                   70.799246     ...               4752.971680   \n",
       "5                  186.223285     ...               7981.845703   \n",
       "\n",
       "          PS_HH_BLKGRP_ALRND   PS_B01001e1  PS_B11001e1  PS_B08301e1  \\\n",
       "OBJECTID                                                               \n",
       "1                8365.437500  11049.640625  5791.404785  6460.674805   \n",
       "2                4022.760010   8285.339844  3626.171387  5166.641113   \n",
       "3                1881.152710   7119.014648  1994.968506  3319.922363   \n",
       "4                1443.358765   5364.205078  1576.304443  2900.467041   \n",
       "5                2237.658447   7120.717285  2132.366699  3871.077148   \n",
       "\n",
       "          PS_B08301e18  PS_B08301e19  PS_B25044e1  PS_B25044e10  PS_B25044e3  \n",
       "OBJECTID                                                                      \n",
       "1           328.934082   1452.118164  5791.404785   1854.352051   280.775940  \n",
       "2           403.354156    410.178406  3626.171387    320.021942    66.299080  \n",
       "3             2.154067     91.777306  1994.968506    213.593430   143.926300  \n",
       "4            46.356518    356.708252  1576.304443    254.681091    20.526436  \n",
       "5            53.870941    173.980942  2132.366699     96.113113    33.129517  \n",
       "\n",
       "[5 rows x 64 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stops_fields = [i.name for i in arcpy.ListFields(stops_data) if i.type not in (\"OID\", \"Geometry\")]\n",
    "df = arcgis_table_to_df(stops_data,stops_fields)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>BART</th>\n",
       "      <th>N_S_ORDER</th>\n",
       "      <th>STATUS</th>\n",
       "      <th>index</th>\n",
       "      <th>FF_Bike_Raw_Min_FF</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_JBS</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_POP</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_HH</th>\n",
       "      <th>FF_Bike_LTS_Walk_FF</th>\n",
       "      <th>ACCESS_Bike_LTS_Walk_JBS</th>\n",
       "      <th>...</th>\n",
       "      <th>PS_POP_BLKGRP_ALRND</th>\n",
       "      <th>PS_HH_BLKGRP_ALRND</th>\n",
       "      <th>PS_B01001e1</th>\n",
       "      <th>PS_B11001e1</th>\n",
       "      <th>PS_B08301e1</th>\n",
       "      <th>PS_B08301e18</th>\n",
       "      <th>PS_B08301e19</th>\n",
       "      <th>PS_B25044e1</th>\n",
       "      <th>PS_B25044e10</th>\n",
       "      <th>PS_B25044e3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>38.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.105263</td>\n",
       "      <td>19.500000</td>\n",
       "      <td>0.868421</td>\n",
       "      <td>18.500000</td>\n",
       "      <td>415.017914</td>\n",
       "      <td>34680.965542</td>\n",
       "      <td>36976.467837</td>\n",
       "      <td>14247.269853</td>\n",
       "      <td>276.091249</td>\n",
       "      <td>22978.245982</td>\n",
       "      <td>...</td>\n",
       "      <td>6660.213646</td>\n",
       "      <td>2614.441793</td>\n",
       "      <td>6520.950365</td>\n",
       "      <td>2537.196889</td>\n",
       "      <td>3502.853619</td>\n",
       "      <td>132.559213</td>\n",
       "      <td>266.674541</td>\n",
       "      <td>2537.196889</td>\n",
       "      <td>284.899261</td>\n",
       "      <td>55.319307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.311012</td>\n",
       "      <td>11.113055</td>\n",
       "      <td>1.234001</td>\n",
       "      <td>11.113055</td>\n",
       "      <td>152.482854</td>\n",
       "      <td>40403.692868</td>\n",
       "      <td>18612.097079</td>\n",
       "      <td>9231.806037</td>\n",
       "      <td>134.168847</td>\n",
       "      <td>31119.314966</td>\n",
       "      <td>...</td>\n",
       "      <td>3047.232089</td>\n",
       "      <td>1570.660610</td>\n",
       "      <td>2598.592172</td>\n",
       "      <td>1276.397973</td>\n",
       "      <td>1526.195483</td>\n",
       "      <td>140.190093</td>\n",
       "      <td>489.982871</td>\n",
       "      <td>1276.397973</td>\n",
       "      <td>464.713448</td>\n",
       "      <td>70.844941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>89.127951</td>\n",
       "      <td>1295.799699</td>\n",
       "      <td>4103.860558</td>\n",
       "      <td>1249.571704</td>\n",
       "      <td>30.371018</td>\n",
       "      <td>399.949189</td>\n",
       "      <td>...</td>\n",
       "      <td>1121.132935</td>\n",
       "      <td>265.259766</td>\n",
       "      <td>236.207626</td>\n",
       "      <td>58.696388</td>\n",
       "      <td>141.165436</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000732</td>\n",
       "      <td>58.696388</td>\n",
       "      <td>1.019479</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>10.250000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.250000</td>\n",
       "      <td>313.800752</td>\n",
       "      <td>16992.536989</td>\n",
       "      <td>26204.694970</td>\n",
       "      <td>9267.001723</td>\n",
       "      <td>183.559441</td>\n",
       "      <td>9262.099185</td>\n",
       "      <td>...</td>\n",
       "      <td>4666.104004</td>\n",
       "      <td>1721.583832</td>\n",
       "      <td>5063.377441</td>\n",
       "      <td>1690.401520</td>\n",
       "      <td>2559.117249</td>\n",
       "      <td>29.392113</td>\n",
       "      <td>48.189219</td>\n",
       "      <td>1690.401520</td>\n",
       "      <td>91.598263</td>\n",
       "      <td>13.230458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>19.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>18.500000</td>\n",
       "      <td>423.307925</td>\n",
       "      <td>21682.554801</td>\n",
       "      <td>30208.107795</td>\n",
       "      <td>11067.225298</td>\n",
       "      <td>266.503309</td>\n",
       "      <td>12639.256578</td>\n",
       "      <td>...</td>\n",
       "      <td>6237.207520</td>\n",
       "      <td>2234.932983</td>\n",
       "      <td>6197.038818</td>\n",
       "      <td>2227.485229</td>\n",
       "      <td>2992.486206</td>\n",
       "      <td>60.252720</td>\n",
       "      <td>120.616875</td>\n",
       "      <td>2227.485229</td>\n",
       "      <td>168.862625</td>\n",
       "      <td>33.681007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>28.750000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>27.750000</td>\n",
       "      <td>518.370363</td>\n",
       "      <td>33097.337913</td>\n",
       "      <td>44388.402313</td>\n",
       "      <td>15521.807589</td>\n",
       "      <td>362.066119</td>\n",
       "      <td>20003.046022</td>\n",
       "      <td>...</td>\n",
       "      <td>8444.630615</td>\n",
       "      <td>3166.637085</td>\n",
       "      <td>8139.424316</td>\n",
       "      <td>3053.257141</td>\n",
       "      <td>4561.344360</td>\n",
       "      <td>225.179981</td>\n",
       "      <td>240.976742</td>\n",
       "      <td>3053.257141</td>\n",
       "      <td>271.081062</td>\n",
       "      <td>56.264241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>696.239677</td>\n",
       "      <td>203825.650070</td>\n",
       "      <td>80442.088298</td>\n",
       "      <td>41191.866933</td>\n",
       "      <td>543.665366</td>\n",
       "      <td>163234.149065</td>\n",
       "      <td>...</td>\n",
       "      <td>15357.574219</td>\n",
       "      <td>8365.437500</td>\n",
       "      <td>12277.284180</td>\n",
       "      <td>6528.523438</td>\n",
       "      <td>6771.132324</td>\n",
       "      <td>456.082886</td>\n",
       "      <td>2755.272461</td>\n",
       "      <td>6528.523438</td>\n",
       "      <td>2404.553955</td>\n",
       "      <td>326.985992</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 63 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            BART  N_S_ORDER     STATUS      index  FF_Bike_Raw_Min_FF  \\\n",
       "count  38.000000  38.000000  38.000000  38.000000           38.000000   \n",
       "mean    0.105263  19.500000   0.868421  18.500000          415.017914   \n",
       "std     0.311012  11.113055   1.234001  11.113055          152.482854   \n",
       "min     0.000000   1.000000   0.000000   0.000000           89.127951   \n",
       "25%     0.000000  10.250000   0.000000   9.250000          313.800752   \n",
       "50%     0.000000  19.500000   0.000000  18.500000          423.307925   \n",
       "75%     0.000000  28.750000   2.000000  27.750000          518.370363   \n",
       "max     1.000000  38.000000   4.000000  37.000000          696.239677   \n",
       "\n",
       "       ACCESS_Bike_Raw_Min_JBS  ACCESS_Bike_Raw_Min_POP  \\\n",
       "count                38.000000                38.000000   \n",
       "mean              34680.965542             36976.467837   \n",
       "std               40403.692868             18612.097079   \n",
       "min                1295.799699              4103.860558   \n",
       "25%               16992.536989             26204.694970   \n",
       "50%               21682.554801             30208.107795   \n",
       "75%               33097.337913             44388.402313   \n",
       "max              203825.650070             80442.088298   \n",
       "\n",
       "       ACCESS_Bike_Raw_Min_HH  FF_Bike_LTS_Walk_FF  ACCESS_Bike_LTS_Walk_JBS  \\\n",
       "count               38.000000            38.000000                 38.000000   \n",
       "mean             14247.269853           276.091249              22978.245982   \n",
       "std               9231.806037           134.168847              31119.314966   \n",
       "min               1249.571704            30.371018                399.949189   \n",
       "25%               9267.001723           183.559441               9262.099185   \n",
       "50%              11067.225298           266.503309              12639.256578   \n",
       "75%              15521.807589           362.066119              20003.046022   \n",
       "max              41191.866933           543.665366             163234.149065   \n",
       "\n",
       "          ...       PS_POP_BLKGRP_ALRND  PS_HH_BLKGRP_ALRND   PS_B01001e1  \\\n",
       "count     ...                 38.000000           38.000000     38.000000   \n",
       "mean      ...               6660.213646         2614.441793   6520.950365   \n",
       "std       ...               3047.232089         1570.660610   2598.592172   \n",
       "min       ...               1121.132935          265.259766    236.207626   \n",
       "25%       ...               4666.104004         1721.583832   5063.377441   \n",
       "50%       ...               6237.207520         2234.932983   6197.038818   \n",
       "75%       ...               8444.630615         3166.637085   8139.424316   \n",
       "max       ...              15357.574219         8365.437500  12277.284180   \n",
       "\n",
       "       PS_B11001e1  PS_B08301e1  PS_B08301e18  PS_B08301e19  PS_B25044e1  \\\n",
       "count    38.000000    38.000000     38.000000     38.000000    38.000000   \n",
       "mean   2537.196889  3502.853619    132.559213    266.674541  2537.196889   \n",
       "std    1276.397973  1526.195483    140.190093    489.982871  1276.397973   \n",
       "min      58.696388   141.165436      0.000000      5.000732    58.696388   \n",
       "25%    1690.401520  2559.117249     29.392113     48.189219  1690.401520   \n",
       "50%    2227.485229  2992.486206     60.252720    120.616875  2227.485229   \n",
       "75%    3053.257141  4561.344360    225.179981    240.976742  3053.257141   \n",
       "max    6528.523438  6771.132324    456.082886   2755.272461  6528.523438   \n",
       "\n",
       "       PS_B25044e10  PS_B25044e3  \n",
       "count     38.000000    38.000000  \n",
       "mean     284.899261    55.319307  \n",
       "std      464.713448    70.844941  \n",
       "min        1.019479     0.000000  \n",
       "25%       91.598263    13.230458  \n",
       "50%      168.862625    33.681007  \n",
       "75%      271.081062    56.264241  \n",
       "max     2404.553955   326.985992  \n",
       "\n",
       "[8 rows x 63 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Rename and Clean Data Frame \n",
    "Census & other data needs readable names and needs to be preprocessed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Percent Commute to Work Via Bicycle</th>\n",
       "      <th>Percent Zero Vehicle Households</th>\n",
       "      <th>Bike_Parking_Total</th>\n",
       "      <th>Jobs Within 1/2 Mile</th>\n",
       "      <th>Population Within 1/2 MIle</th>\n",
       "      <th>Daily_Bike_Riders</th>\n",
       "      <th>AM_Bike_Riders</th>\n",
       "      <th>PM_Bike_Riders</th>\n",
       "      <th>ACCESS_Bike_LTS_Hard_JBS</th>\n",
       "      <th>ACCESS_Bike_LTS_Hard_POP</th>\n",
       "      <th>ACCESS_Bike_LTS_Hard_HH</th>\n",
       "      <th>ACCESS_Bike_LTS_Walk_JBS</th>\n",
       "      <th>ACCESS_Bike_LTS_Walk_POP</th>\n",
       "      <th>ACCESS_Bike_LTS_Walk_HH</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_JBS</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_POP</th>\n",
       "      <th>ACCESS_Bike_Raw_Min_HH</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>29.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.034624</td>\n",
       "      <td>0.093755</td>\n",
       "      <td>75.965517</td>\n",
       "      <td>7079.986061</td>\n",
       "      <td>6571.686553</td>\n",
       "      <td>374.068966</td>\n",
       "      <td>153.344828</td>\n",
       "      <td>148.551724</td>\n",
       "      <td>6163.786130</td>\n",
       "      <td>8123.148499</td>\n",
       "      <td>3082.236511</td>\n",
       "      <td>17008.775260</td>\n",
       "      <td>21192.603838</td>\n",
       "      <td>8020.111290</td>\n",
       "      <td>26879.798688</td>\n",
       "      <td>33606.595150</td>\n",
       "      <td>12492.131954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.041126</td>\n",
       "      <td>0.067930</td>\n",
       "      <td>97.311900</td>\n",
       "      <td>6816.577021</td>\n",
       "      <td>2989.189063</td>\n",
       "      <td>605.925970</td>\n",
       "      <td>194.104036</td>\n",
       "      <td>307.768302</td>\n",
       "      <td>11082.798207</td>\n",
       "      <td>11563.209564</td>\n",
       "      <td>4757.597260</td>\n",
       "      <td>20742.905981</td>\n",
       "      <td>12230.073919</td>\n",
       "      <td>5766.835684</td>\n",
       "      <td>28160.626910</td>\n",
       "      <td>15609.034348</td>\n",
       "      <td>7267.502030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.017369</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>303.214417</td>\n",
       "      <td>1121.132935</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>399.949189</td>\n",
       "      <td>1231.528267</td>\n",
       "      <td>351.404031</td>\n",
       "      <td>1295.799699</td>\n",
       "      <td>4103.860558</td>\n",
       "      <td>1249.571704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.010409</td>\n",
       "      <td>0.057725</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>3556.007080</td>\n",
       "      <td>4685.727539</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>10.157112</td>\n",
       "      <td>6.239936</td>\n",
       "      <td>5.717268</td>\n",
       "      <td>8697.225906</td>\n",
       "      <td>14499.425460</td>\n",
       "      <td>5195.512554</td>\n",
       "      <td>17115.861176</td>\n",
       "      <td>25991.145208</td>\n",
       "      <td>9528.871827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.016331</td>\n",
       "      <td>0.082917</td>\n",
       "      <td>43.000000</td>\n",
       "      <td>6023.424316</td>\n",
       "      <td>6149.746094</td>\n",
       "      <td>186.000000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>2118.988237</td>\n",
       "      <td>1952.303185</td>\n",
       "      <td>717.362621</td>\n",
       "      <td>12144.297060</td>\n",
       "      <td>20580.317818</td>\n",
       "      <td>7416.299127</td>\n",
       "      <td>19429.776356</td>\n",
       "      <td>29792.148260</td>\n",
       "      <td>11062.682705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.050913</td>\n",
       "      <td>0.105064</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>7861.391602</td>\n",
       "      <td>8123.459961</td>\n",
       "      <td>425.000000</td>\n",
       "      <td>204.000000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>7415.726017</td>\n",
       "      <td>15167.258059</td>\n",
       "      <td>5722.821203</td>\n",
       "      <td>16961.012942</td>\n",
       "      <td>25215.873124</td>\n",
       "      <td>9479.435229</td>\n",
       "      <td>28621.858478</td>\n",
       "      <td>41386.049757</td>\n",
       "      <td>14404.996185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.158466</td>\n",
       "      <td>0.368672</td>\n",
       "      <td>430.000000</td>\n",
       "      <td>33647.035156</td>\n",
       "      <td>15357.574219</td>\n",
       "      <td>3060.000000</td>\n",
       "      <td>927.000000</td>\n",
       "      <td>1461.000000</td>\n",
       "      <td>52875.828229</td>\n",
       "      <td>52000.007254</td>\n",
       "      <td>22616.948990</td>\n",
       "      <td>108533.307052</td>\n",
       "      <td>57943.928318</td>\n",
       "      <td>27969.887542</td>\n",
       "      <td>153207.748871</td>\n",
       "      <td>78579.951128</td>\n",
       "      <td>38569.412104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Percent Commute to Work Via Bicycle  Percent Zero Vehicle Households  \\\n",
       "count                            29.000000                        29.000000   \n",
       "mean                              0.034624                         0.093755   \n",
       "std                               0.041126                         0.067930   \n",
       "min                               0.000000                         0.017369   \n",
       "25%                               0.010409                         0.057725   \n",
       "50%                               0.016331                         0.082917   \n",
       "75%                               0.050913                         0.105064   \n",
       "max                               0.158466                         0.368672   \n",
       "\n",
       "       Bike_Parking_Total  Jobs Within 1/2 Mile  Population Within 1/2 MIle  \\\n",
       "count           29.000000             29.000000                   29.000000   \n",
       "mean            75.965517           7079.986061                 6571.686553   \n",
       "std             97.311900           6816.577021                 2989.189063   \n",
       "min              0.000000            303.214417                 1121.132935   \n",
       "25%             30.000000           3556.007080                 4685.727539   \n",
       "50%             43.000000           6023.424316                 6149.746094   \n",
       "75%             76.000000           7861.391602                 8123.459961   \n",
       "max            430.000000          33647.035156                15357.574219   \n",
       "\n",
       "       Daily_Bike_Riders  AM_Bike_Riders  PM_Bike_Riders  \\\n",
       "count          29.000000       29.000000       29.000000   \n",
       "mean          374.068966      153.344828      148.551724   \n",
       "std           605.925970      194.104036      307.768302   \n",
       "min             0.000000        0.000000        0.000000   \n",
       "25%            70.000000       28.000000        9.000000   \n",
       "50%           186.000000       94.000000       38.000000   \n",
       "75%           425.000000      204.000000       89.000000   \n",
       "max          3060.000000      927.000000     1461.000000   \n",
       "\n",
       "       ACCESS_Bike_LTS_Hard_JBS  ACCESS_Bike_LTS_Hard_POP  \\\n",
       "count                 29.000000                 29.000000   \n",
       "mean                6163.786130               8123.148499   \n",
       "std                11082.798207              11563.209564   \n",
       "min                    0.000000                  0.000000   \n",
       "25%                   10.157112                  6.239936   \n",
       "50%                 2118.988237               1952.303185   \n",
       "75%                 7415.726017              15167.258059   \n",
       "max                52875.828229              52000.007254   \n",
       "\n",
       "       ACCESS_Bike_LTS_Hard_HH  ACCESS_Bike_LTS_Walk_JBS  \\\n",
       "count                29.000000                 29.000000   \n",
       "mean               3082.236511              17008.775260   \n",
       "std                4757.597260              20742.905981   \n",
       "min                   0.000000                399.949189   \n",
       "25%                   5.717268               8697.225906   \n",
       "50%                 717.362621              12144.297060   \n",
       "75%                5722.821203              16961.012942   \n",
       "max               22616.948990             108533.307052   \n",
       "\n",
       "       ACCESS_Bike_LTS_Walk_POP  ACCESS_Bike_LTS_Walk_HH  \\\n",
       "count                 29.000000                29.000000   \n",
       "mean               21192.603838              8020.111290   \n",
       "std                12230.073919              5766.835684   \n",
       "min                 1231.528267               351.404031   \n",
       "25%                14499.425460              5195.512554   \n",
       "50%                20580.317818              7416.299127   \n",
       "75%                25215.873124              9479.435229   \n",
       "max                57943.928318             27969.887542   \n",
       "\n",
       "       ACCESS_Bike_Raw_Min_JBS  ACCESS_Bike_Raw_Min_POP  \\\n",
       "count                29.000000                29.000000   \n",
       "mean              26879.798688             33606.595150   \n",
       "std               28160.626910             15609.034348   \n",
       "min                1295.799699              4103.860558   \n",
       "25%               17115.861176             25991.145208   \n",
       "50%               19429.776356             29792.148260   \n",
       "75%               28621.858478             41386.049757   \n",
       "max              153207.748871             78579.951128   \n",
       "\n",
       "       ACCESS_Bike_Raw_Min_HH  \n",
       "count               29.000000  \n",
       "mean             12492.131954  \n",
       "std               7267.502030  \n",
       "min               1249.571704  \n",
       "25%               9528.871827  \n",
       "50%              11062.682705  \n",
       "75%              14404.996185  \n",
       "max              38569.412104  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_df = df.dropna(subset= [\"Daily_Bike_Riders\"]).copy()\n",
    "reg_df = reg_df.fillna(0)\n",
    "control_fields = [\"Percent Commute to Work Via Bicycle\",\"Percent Zero Vehicle Households\",\"Bike_Parking_Total\",\"Jobs Within 1/2 Mile\",\"Population Within 1/2 MIle\",]\n",
    "control_fields_no_jobs_pop = [\"Percent Commute to Work Via Bicycle\",\"Percent Zero Vehicle Households\",\"Bike_Parking_Total\"]\n",
    "hard_access_fields = [\"ACCESS_Bike_LTS_Hard_JBS\",\"ACCESS_Bike_LTS_Hard_POP\",\"ACCESS_Bike_LTS_Hard_HH\"]\n",
    "soft_access_fields = [\"ACCESS_Bike_LTS_Walk_JBS\",\"ACCESS_Bike_LTS_Walk_POP\",\"ACCESS_Bike_LTS_Walk_HH\"]\n",
    "bike_access_fields = [\"ACCESS_Bike_Raw_Min_JBS\",\"ACCESS_Bike_Raw_Min_POP\",\"ACCESS_Bike_Raw_Min_HH\"]\n",
    "dependents = [\"Daily_Bike_Riders\",\"AM_Bike_Riders\",\"PM_Bike_Riders\"]\n",
    "all_independents = control_fields +hard_access_fields + soft_access_fields + bike_access_fields\n",
    "reg_df[\"Percent Commute to Work Via Bicycle\"] = reg_df[\"PS_B08301e18\"]/reg_df[\"PS_B08301e1\"]\n",
    "reg_df[\"Percent Zero Vehicle Households\"] = (reg_df[\"PS_B25044e10\"]+reg_df[\"PS_B25044e3\"])/reg_df[\"PS_B25044e1\"]\n",
    "rename_dict = {\"PS_SUM_JBc000\":\"Jobs Within 1/2 Mile\",\"PS_POP_BLKGRP_ALRND\":\"Population Within 1/2 MIle\"}\n",
    "reg_df = reg_df.rename(rename_dict,axis=1)\n",
    "data_features = control_fields+ dependents + hard_access_fields + soft_access_fields + bike_access_fields\n",
    "reg_df[data_features].describe()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exploratory Charting and Correlation Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 72 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate Pair Plot\n",
    "pair_plot=sns.pairplot(reg_df[dependents + control_fields])\n",
    "plt.show(pair_plot)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Choice of Independents\n",
    "Correlation analysis and scatter plots indicate that soft-lts accessibility is a better fit as a predictor of ridership, but it is very close to raw bicycle access. Scatter plots of soft access show a close relationship. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate Correlation Matrix\n",
    "data_for_correlation_computation=reg_df[data_features].copy()\n",
    "\n",
    "corr_df=data_for_correlation_computation.corr()\n",
    "#Generate Mask for Upper Triangle\n",
    "mask= np.zeros_like(corr_df, dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "#Generate correlation heat map\n",
    "with sns.axes_style(\"ticks\"):\n",
    "    sns.set(rc={'figure.figsize':(10,10)})\n",
    "    axplot=sns.heatmap(corr_df,mask=mask, annot=True, linewidths=.6,linecolor=\"white\",cmap=\"seismic\",vmin=-1,vmax=1)\n",
    "    plt.show(axplot)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate Correlation Matrix\n",
    "data_for_correlation_computation=reg_df[data_features].copy()[reg_df[\"Name\"]!=\"4th and King\"] # Examine Relationship with outlier Out\n",
    "\n",
    "corr_df=data_for_correlation_computation.corr()\n",
    "#Generate Mask for Upper Triangle\n",
    "mask= np.zeros_like(corr_df, dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "#Generate correlation heat map\n",
    "with sns.axes_style(\"ticks\"):\n",
    "    sns.set(rc={'figure.figsize':(10,10)})\n",
    "    axplot=sns.heatmap(corr_df,mask=mask, annot=True, linewidths=.6,linecolor=\"white\",cmap=\"seismic\",vmin=-1,vmax=1)\n",
    "    plt.show(axplot)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 42 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate Pair Plot\n",
    "pair_plot=sns.pairplot(reg_df[dependents + soft_access_fields],kind=\"reg\")\n",
    "plt.show(pair_plot)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regression Analysis\n",
    "Small sample sizes can be a large reason for large adjusted R^2's. \n",
    "* We want a model that behaves as expected in terms of the direction of coefficients (positive coefficients based on the positive correlations identified above).\n",
    "* We want sparse model that explains a good portion of the observed ridership.\n",
    "* We want at least 1-2 control variables but as our variables have a high degree of collinearity (and circular logic in the case of bike parking) we need to test these variables in the context of TCRP 153 without being too beholden to its results. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "independents = control_fields\n",
    "daily_ridership = dependents[0]\n",
    "am_ridership = dependents[1]\n",
    "pm_ridership = dependents[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.917</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>        <th>  Adj. R-squared:    </th> <td>   0.900</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   52.94</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>3.51e-12</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:04</td>      <th>  Log-Likelihood:    </th> <td> -195.20</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   400.4</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    24</td>       <th>  BIC:               </th> <td>   407.2</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     5</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                   <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Commute to Work Via Bicycle</th> <td>-2229.7274</td> <td> 1171.718</td> <td>   -1.903</td> <td> 0.069</td> <td>-4648.034</td> <td>  188.579</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th>     <td>-1463.7205</td> <td> 1128.685</td> <td>   -1.297</td> <td> 0.207</td> <td>-3793.211</td> <td>  865.770</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Bike_Parking_Total</th>                  <td>    3.6414</td> <td>    0.906</td> <td>    4.019</td> <td> 0.001</td> <td>    1.771</td> <td>    5.512</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Jobs Within 1/2 Mile</th>                <td>    0.0473</td> <td>    0.016</td> <td>    3.006</td> <td> 0.006</td> <td>    0.015</td> <td>    0.080</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Population Within 1/2 MIle</th>          <td>    0.0031</td> <td>    0.011</td> <td>    0.272</td> <td> 0.788</td> <td>   -0.020</td> <td>    0.026</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 5.393</td> <th>  Durbin-Watson:     </th> <td>   1.711</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.067</td> <th>  Jarque-Bera (JB):  </th> <td>   3.772</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.834</td> <th>  Prob(JB):          </th> <td>   0.152</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.585</td> <th>  Cond. No.          </th> <td>3.55e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 3.55e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.917\n",
       "Model:                            OLS   Adj. R-squared:                  0.900\n",
       "Method:                 Least Squares   F-statistic:                     52.94\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.51e-12\n",
       "Time:                        18:06:04   Log-Likelihood:                -195.20\n",
       "No. Observations:                  29   AIC:                             400.4\n",
       "Df Residuals:                      24   BIC:                             407.2\n",
       "Df Model:                           5                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=======================================================================================================\n",
       "                                          coef    std err          t      P>|t|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------------------------------\n",
       "Percent Commute to Work Via Bicycle -2229.7274   1171.718     -1.903      0.069   -4648.034     188.579\n",
       "Percent Zero Vehicle Households     -1463.7205   1128.685     -1.297      0.207   -3793.211     865.770\n",
       "Bike_Parking_Total                      3.6414      0.906      4.019      0.001       1.771       5.512\n",
       "Jobs Within 1/2 Mile                    0.0473      0.016      3.006      0.006       0.015       0.080\n",
       "Population Within 1/2 MIle              0.0031      0.011      0.272      0.788      -0.020       0.026\n",
       "==============================================================================\n",
       "Omnibus:                        5.393   Durbin-Watson:                   1.711\n",
       "Prob(Omnibus):                  0.067   Jarque-Bera (JB):                3.772\n",
       "Skew:                           0.834   Prob(JB):                        0.152\n",
       "Kurtosis:                       3.585   Cond. No.                     3.55e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 3.55e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_mod = sm.OLS(reg_df[daily_ridership],reg_df[independents])\n",
    "daily_res = daily_mod.fit() \n",
    "daily_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Percent Commute to Work Via Bicycle',\n",
       " 'Percent Zero Vehicle Households',\n",
       " 'Bike_Parking_Total',\n",
       " 'ACCESS_Bike_LTS_Walk_JBS',\n",
       " 'ACCESS_Bike_LTS_Walk_POP',\n",
       " 'ACCESS_Bike_LTS_Walk_HH']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "independents_access = control_fields_no_jobs_pop + soft_access_fields \n",
    "independents_access"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.948</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>        <th>  Adj. R-squared:    </th> <td>   0.934</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   69.77</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>1.34e-13</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:04</td>      <th>  Log-Likelihood:    </th> <td> -188.41</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   388.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    23</td>       <th>  BIC:               </th> <td>   397.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     6</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                   <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Commute to Work Via Bicycle</th> <td>-1487.9407</td> <td> 1164.480</td> <td>   -1.278</td> <td> 0.214</td> <td>-3896.852</td> <td>  920.970</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th>     <td> -741.8483</td> <td>  968.870</td> <td>   -0.766</td> <td> 0.452</td> <td>-2746.108</td> <td> 1262.411</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Bike_Parking_Total</th>                  <td>    4.0925</td> <td>    0.643</td> <td>    6.367</td> <td> 0.000</td> <td>    2.763</td> <td>    5.422</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>            <td>    0.0220</td> <td>    0.009</td> <td>    2.539</td> <td> 0.018</td> <td>    0.004</td> <td>    0.040</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>            <td>    0.0130</td> <td>    0.027</td> <td>    0.487</td> <td> 0.631</td> <td>   -0.042</td> <td>    0.068</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_HH</th>             <td>   -0.0570</td> <td>    0.078</td> <td>   -0.732</td> <td> 0.472</td> <td>   -0.218</td> <td>    0.104</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 2.457</td> <th>  Durbin-Watson:     </th> <td>   2.189</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.293</td> <th>  Jarque-Bera (JB):  </th> <td>   1.979</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.630</td> <th>  Prob(JB):          </th> <td>   0.372</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.779</td> <th>  Cond. No.          </th> <td>1.43e+06</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 1.43e+06. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.948\n",
       "Model:                            OLS   Adj. R-squared:                  0.934\n",
       "Method:                 Least Squares   F-statistic:                     69.77\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           1.34e-13\n",
       "Time:                        18:06:04   Log-Likelihood:                -188.41\n",
       "No. Observations:                  29   AIC:                             388.8\n",
       "Df Residuals:                      23   BIC:                             397.0\n",
       "Df Model:                           6                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=======================================================================================================\n",
       "                                          coef    std err          t      P>|t|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------------------------------\n",
       "Percent Commute to Work Via Bicycle -1487.9407   1164.480     -1.278      0.214   -3896.852     920.970\n",
       "Percent Zero Vehicle Households      -741.8483    968.870     -0.766      0.452   -2746.108    1262.411\n",
       "Bike_Parking_Total                      4.0925      0.643      6.367      0.000       2.763       5.422\n",
       "ACCESS_Bike_LTS_Walk_JBS                0.0220      0.009      2.539      0.018       0.004       0.040\n",
       "ACCESS_Bike_LTS_Walk_POP                0.0130      0.027      0.487      0.631      -0.042       0.068\n",
       "ACCESS_Bike_LTS_Walk_HH                -0.0570      0.078     -0.732      0.472      -0.218       0.104\n",
       "==============================================================================\n",
       "Omnibus:                        2.457   Durbin-Watson:                   2.189\n",
       "Prob(Omnibus):                  0.293   Jarque-Bera (JB):                1.979\n",
       "Skew:                           0.630   Prob(JB):                        0.372\n",
       "Kurtosis:                       2.779   Cond. No.                     1.43e+06\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 1.43e+06. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_mod = sm.OLS(reg_df[daily_ridership],reg_df[independents_access])\n",
    "daily_res = daily_mod.fit() \n",
    "daily_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 48.8%, #d65f5f 48.8%, #d65f5f 55.0%, transparent 55.0%);\n",
       "        }    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 46.7%, #d65f5f 46.7%, #d65f5f 55.0%, transparent 55.0%);\n",
       "        }    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 55.0%, #5fba7d 55.0%, #5fba7d 92.5%, transparent 92.5%);\n",
       "        }    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 55.0%, #5fba7d 55.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 55.0%, #5fba7d 55.0%, #5fba7d 88.2%, transparent 88.2%);\n",
       "        }    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #d65f5f 0.0%, #d65f5f 55.0%, transparent 55.0%);\n",
       "        }    #T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 55.0%, transparent 55.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Commute to Work Via Bicycle</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col2\" class=\"data row0 col2\" >-1487.94</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0346245</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.214077</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row0_col5\" class=\"data row0 col5\" >-0.137726</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col1\" class=\"data row1 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col2\" class=\"data row1 col2\" >-741.848</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col3\" class=\"data row1 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.451651</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row1_col5\" class=\"data row1 col5\" >-0.185933</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col0\" class=\"data row2 col0\" >Independent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col1\" class=\"data row2 col1\" >Bike_Parking_Total</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col2\" class=\"data row2 col2\" >4.09254</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col3\" class=\"data row2 col3\" >75.9655</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col4\" class=\"data row2 col4\" >1.69736e-06</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row2_col5\" class=\"data row2 col5\" >0.831108</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row3\" class=\"row_heading level0 row3\" >3</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col0\" class=\"data row3 col0\" >Independent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col1\" class=\"data row3 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col2\" class=\"data row3 col2\" >0.0219628</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col3\" class=\"data row3 col3\" >17008.8</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col4\" class=\"data row3 col4\" >0.0183247</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row3_col5\" class=\"data row3 col5\" >0.998641</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row4\" class=\"row_heading level0 row4\" >4</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col0\" class=\"data row4 col0\" >Independent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col1\" class=\"data row4 col1\" >ACCESS_Bike_LTS_Walk_POP</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col2\" class=\"data row4 col2\" >0.0130017</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col3\" class=\"data row4 col3\" >21192.6</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col4\" class=\"data row4 col4\" >0.631181</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row4_col5\" class=\"data row4 col5\" >0.736602</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row5\" class=\"row_heading level0 row5\" >5</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col0\" class=\"data row5 col0\" >Independent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col1\" class=\"data row5 col1\" >ACCESS_Bike_LTS_Walk_HH</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col2\" class=\"data row5 col2\" >-0.0570304</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col3\" class=\"data row5 col3\" >8020.11</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col4\" class=\"data row5 col4\" >0.471715</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row5_col5\" class=\"data row5 col5\" >-1.22274</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14level0_row6\" class=\"row_heading level0 row6\" >6</th> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col0\" class=\"data row6 col0\" >Dependent</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col1\" class=\"data row6 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col2\" class=\"data row6 col2\" >nan</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col3\" class=\"data row6 col3\" >374.069</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col4\" class=\"data row6 col4\" >nan</td> \n",
       "        <td id=\"T_c29e2380_29b3_11e9_9d64_10f005ecad14row6_col5\" class=\"data row6 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x21308905da0>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_res,reg_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.940</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>        <th>  Adj. R-squared:    </th> <td>   0.931</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   98.57</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>6.29e-15</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -190.38</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   388.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    25</td>       <th>  BIC:               </th> <td>   394.2</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     4</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                   <td></td>                      <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th>     <td>-1016.5215</td> <td>  652.149</td> <td>   -1.559</td> <td> 0.132</td> <td>-2359.647</td> <td>  326.604</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Bike_Parking_Total</th>                  <td>    4.4632</td> <td>    0.599</td> <td>    7.456</td> <td> 0.000</td> <td>    3.230</td> <td>    5.696</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Commute to Work Via Bicycle</th> <td>-1810.0900</td> <td>  923.149</td> <td>   -1.961</td> <td> 0.061</td> <td>-3711.351</td> <td>   91.171</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>            <td>    0.0129</td> <td>    0.003</td> <td>    4.819</td> <td> 0.000</td> <td>    0.007</td> <td>    0.018</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 1.806</td> <th>  Durbin-Watson:     </th> <td>   2.181</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.405</td> <th>  Jarque-Bera (JB):  </th> <td>   1.517</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.538</td> <th>  Prob(JB):          </th> <td>   0.468</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.688</td> <th>  Cond. No.          </th> <td>7.14e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 7.14e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.940\n",
       "Model:                            GLS   Adj. R-squared:                  0.931\n",
       "Method:                 Least Squares   F-statistic:                     98.57\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           6.29e-15\n",
       "Time:                        18:06:05   Log-Likelihood:                -190.38\n",
       "No. Observations:                  29   AIC:                             388.8\n",
       "Df Residuals:                      25   BIC:                             394.2\n",
       "Df Model:                           4                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=======================================================================================================\n",
       "                                          coef    std err          t      P>|t|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households     -1016.5215    652.149     -1.559      0.132   -2359.647     326.604\n",
       "Bike_Parking_Total                      4.4632      0.599      7.456      0.000       3.230       5.696\n",
       "Percent Commute to Work Via Bicycle -1810.0900    923.149     -1.961      0.061   -3711.351      91.171\n",
       "ACCESS_Bike_LTS_Walk_JBS                0.0129      0.003      4.819      0.000       0.007       0.018\n",
       "==============================================================================\n",
       "Omnibus:                        1.806   Durbin-Watson:                   2.181\n",
       "Prob(Omnibus):                  0.405   Jarque-Bera (JB):                1.517\n",
       "Skew:                           0.538   Prob(JB):                        0.468\n",
       "Kurtosis:                       2.688   Cond. No.                     7.14e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 7.14e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = [\"Percent Zero Vehicle Households\",\"Bike_Parking_Total\" ,\"Percent Commute to Work Via Bicycle\",\"ACCESS_Bike_LTS_Walk_JBS\"]\n",
    "daily_mod = sm.GLS(reg_df[daily_ridership],reg_df[features])\n",
    "daily_res = daily_mod.fit()\n",
    "daily_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "job_control_features = [\"Jobs Within 1/2 Mile\",\"ACCESS_Bike_LTS_Walk_JBS\"]\n",
    "pop_features = [\"Percent Zero Vehicle Households\" ,\"ACCESS_Bike_LTS_Walk_POP\"]\n",
    "job_features = [\"Percent Zero Vehicle Households\",\"ACCESS_Bike_LTS_Walk_JBS\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.875</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>        <th>  Adj. R-squared:    </th> <td>   0.866</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   94.76</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>6.23e-13</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -201.07</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   406.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>       <th>  BIC:               </th> <td>   408.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Jobs Within 1/2 Mile</th>     <td>    0.0439</td> <td>    0.011</td> <td>    4.182</td> <td> 0.000</td> <td>    0.022</td> <td>    0.065</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th> <td>    0.0093</td> <td>    0.004</td> <td>    2.422</td> <td> 0.022</td> <td>    0.001</td> <td>    0.017</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 6.808</td> <th>  Durbin-Watson:     </th> <td>   1.859</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.033</td> <th>  Jarque-Bera (JB):  </th> <td>   5.492</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.683</td> <th>  Prob(JB):          </th> <td>  0.0642</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.636</td> <th>  Cond. No.          </th> <td>    6.47</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.875\n",
       "Model:                            GLS   Adj. R-squared:                  0.866\n",
       "Method:                 Least Squares   F-statistic:                     94.76\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           6.23e-13\n",
       "Time:                        18:06:05   Log-Likelihood:                -201.07\n",
       "No. Observations:                  29   AIC:                             406.1\n",
       "Df Residuals:                      27   BIC:                             408.9\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "Jobs Within 1/2 Mile         0.0439      0.011      4.182      0.000       0.022       0.065\n",
       "ACCESS_Bike_LTS_Walk_JBS     0.0093      0.004      2.422      0.022       0.001       0.017\n",
       "==============================================================================\n",
       "Omnibus:                        6.808   Durbin-Watson:                   1.859\n",
       "Prob(Omnibus):                  0.033   Jarque-Bera (JB):                5.492\n",
       "Skew:                           0.683   Prob(JB):                       0.0642\n",
       "Kurtosis:                       4.636   Cond. No.                         6.47\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_mod = sm.GLS(reg_df[daily_ridership],reg_df[job_control_features])\n",
    "daily_res = daily_mod.fit() \n",
    "daily_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 51.1%, transparent 51.1%);\n",
       "        }    #T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col1\" class=\"data row0 col1\" >Jobs Within 1/2 Mile</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col2\" class=\"data row0 col2\" >0.0439387</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col3\" class=\"data row0 col3\" >7079.99</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.000273099</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.831626</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.00934216</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.0224327</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.424784</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col1\" class=\"data row2 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col3\" class=\"data row2 col3\" >374.069</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c2ab87cc_29b3_11e9_8f8c_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x21308a0a5c0>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_res,reg_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.668</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>        <th>  Adj. R-squared:    </th> <td>   0.643</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   27.15</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>3.44e-07</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -215.28</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   434.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>       <th>  BIC:               </th> <td>   437.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td> 4551.1677</td> <td> 1366.081</td> <td>    3.332</td> <td> 0.003</td> <td> 1748.202</td> <td> 7354.133</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>    0.0024</td> <td>    0.006</td> <td>    0.367</td> <td> 0.716</td> <td>   -0.011</td> <td>    0.016</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>16.446</td> <th>  Durbin-Watson:     </th> <td>   1.182</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  20.802</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.372</td> <th>  Prob(JB):          </th> <td>3.04e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 6.113</td> <th>  Cond. No.          </th> <td>4.27e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.27e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.668\n",
       "Model:                            GLS   Adj. R-squared:                  0.643\n",
       "Method:                 Least Squares   F-statistic:                     27.15\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.44e-07\n",
       "Time:                        18:06:05   Log-Likelihood:                -215.28\n",
       "No. Observations:                  29   AIC:                             434.6\n",
       "Df Residuals:                      27   BIC:                             437.3\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households  4551.1677   1366.081      3.332      0.003    1748.202    7354.133\n",
       "ACCESS_Bike_LTS_Walk_POP            0.0024      0.006      0.367      0.716      -0.011       0.016\n",
       "==============================================================================\n",
       "Omnibus:                       16.446   Durbin-Watson:                   1.182\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               20.802\n",
       "Skew:                           1.372   Prob(JB):                     3.04e-05\n",
       "Kurtosis:                       6.113   Cond. No.                     4.27e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.27e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_mod = sm.GLS(reg_df[daily_ridership],reg_df[pop_features])\n",
    "daily_res = daily_mod.fit()\n",
    "daily_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>AM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.699</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>       <th>  Adj. R-squared:    </th> <td>   0.677</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   31.34</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>9.17e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -183.25</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   370.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>      <th>  BIC:               </th> <td>   373.2</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>  824.0184</td> <td>  452.731</td> <td>    1.820</td> <td> 0.080</td> <td> -104.908</td> <td> 1752.945</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>    0.0048</td> <td>    0.002</td> <td>    2.241</td> <td> 0.033</td> <td>    0.000</td> <td>    0.009</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>13.587</td> <th>  Durbin-Watson:     </th> <td>   1.706</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.001</td> <th>  Jarque-Bera (JB):  </th> <td>  13.369</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.377</td> <th>  Prob(JB):          </th> <td> 0.00125</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.865</td> <th>  Cond. No.          </th> <td>4.27e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.27e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         AM_Bike_Riders   R-squared:                       0.699\n",
       "Model:                            GLS   Adj. R-squared:                  0.677\n",
       "Method:                 Least Squares   F-statistic:                     31.34\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           9.17e-08\n",
       "Time:                        18:06:05   Log-Likelihood:                -183.25\n",
       "No. Observations:                  29   AIC:                             370.5\n",
       "Df Residuals:                      27   BIC:                             373.2\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households   824.0184    452.731      1.820      0.080    -104.908    1752.945\n",
       "ACCESS_Bike_LTS_Walk_POP            0.0048      0.002      2.241      0.033       0.000       0.009\n",
       "==============================================================================\n",
       "Omnibus:                       13.587   Durbin-Watson:                   1.706\n",
       "Prob(Omnibus):                  0.001   Jarque-Bera (JB):               13.369\n",
       "Skew:                           1.377   Prob(JB):                      0.00125\n",
       "Kurtosis:                       4.865   Cond. No.                     4.27e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.27e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "am_mod = sm.GLS(reg_df[am_ridership],reg_df[pop_features])\n",
    "am_res = am_mod.fit()\n",
    "am_res.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Observation - Removing San Francisco \n",
    "When removing an outlier for all variables but one of the few high density land use context from the analysis, job access is still significant when looking  at one variable, but coefficient magnitude drops. This outlier could have some impact on the results. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.668</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>        <th>  Adj. R-squared:    </th> <td>   0.643</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   27.15</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>3.44e-07</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -215.28</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   434.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>       <th>  BIC:               </th> <td>   437.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td> 4551.1677</td> <td> 1366.081</td> <td>    3.332</td> <td> 0.003</td> <td> 1748.202</td> <td> 7354.133</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>    0.0024</td> <td>    0.006</td> <td>    0.367</td> <td> 0.716</td> <td>   -0.011</td> <td>    0.016</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>16.446</td> <th>  Durbin-Watson:     </th> <td>   1.182</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  20.802</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.372</td> <th>  Prob(JB):          </th> <td>3.04e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 6.113</td> <th>  Cond. No.          </th> <td>4.27e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.27e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.668\n",
       "Model:                            GLS   Adj. R-squared:                  0.643\n",
       "Method:                 Least Squares   F-statistic:                     27.15\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.44e-07\n",
       "Time:                        18:06:05   Log-Likelihood:                -215.28\n",
       "No. Observations:                  29   AIC:                             434.6\n",
       "Df Residuals:                      27   BIC:                             437.3\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households  4551.1677   1366.081      3.332      0.003    1748.202    7354.133\n",
       "ACCESS_Bike_LTS_Walk_POP            0.0024      0.006      0.367      0.716      -0.011       0.016\n",
       "==============================================================================\n",
       "Omnibus:                       16.446   Durbin-Watson:                   1.182\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               20.802\n",
       "Skew:                           1.372   Prob(JB):                     3.04e-05\n",
       "Kurtosis:                       6.113   Cond. No.                     4.27e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.27e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_final_job_mod_no_San_Francisco = sm.GLS(reg_df[daily_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[job_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "daily_final_job_res_no_San_Francisco = daily_final_job_mod_no_San_Francisco.fit()\n",
    "daily_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.542</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>        <th>  Adj. R-squared:    </th> <td>   0.507</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   15.41</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>3.85e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -198.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>       <th>  AIC:               </th> <td>   400.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>       <th>  BIC:               </th> <td>   402.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td> 1783.2444</td> <td> 1093.691</td> <td>    1.630</td> <td> 0.115</td> <td> -464.870</td> <td> 4031.358</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>    0.0070</td> <td>    0.005</td> <td>    1.513</td> <td> 0.142</td> <td>   -0.003</td> <td>    0.017</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>15.482</td> <th>  Durbin-Watson:     </th> <td>   1.211</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  16.496</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.488</td> <th>  Prob(JB):          </th> <td>0.000262</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 5.297</td> <th>  Cond. No.          </th> <td>4.35e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.35e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.542\n",
       "Model:                            GLS   Adj. R-squared:                  0.507\n",
       "Method:                 Least Squares   F-statistic:                     15.41\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.85e-05\n",
       "Time:                        18:06:05   Log-Likelihood:                -198.01\n",
       "No. Observations:                  28   AIC:                             400.0\n",
       "Df Residuals:                      26   BIC:                             402.7\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households  1783.2444   1093.691      1.630      0.115    -464.870    4031.358\n",
       "ACCESS_Bike_LTS_Walk_POP            0.0070      0.005      1.513      0.142      -0.003       0.017\n",
       "==============================================================================\n",
       "Omnibus:                       15.482   Durbin-Watson:                   1.211\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               16.496\n",
       "Skew:                           1.488   Prob(JB):                     0.000262\n",
       "Kurtosis:                       5.297   Cond. No.                     4.35e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.35e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_final_pop_mod_no_San_Francisco = sm.GLS(reg_df[daily_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[pop_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "daily_final_pop_res_no_San_Francisco = daily_final_pop_mod_no_San_Francisco.fit()\n",
    "daily_final_pop_res_no_San_Francisco.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.706</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>       <th>  Adj. R-squared:    </th> <td>   0.684</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   32.37</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>6.74e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -192.19</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   388.4</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>      <th>  BIC:               </th> <td>   391.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>  654.3725</td> <td>  588.346</td> <td>    1.112</td> <td> 0.276</td> <td> -552.813</td> <td> 1861.558</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>        <td>    0.0081</td> <td>    0.003</td> <td>    3.190</td> <td> 0.004</td> <td>    0.003</td> <td>    0.013</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>13.323</td> <th>  Durbin-Watson:     </th> <td>   1.616</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.001</td> <th>  Jarque-Bera (JB):  </th> <td>  22.867</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.829</td> <th>  Prob(JB):          </th> <td>1.08e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 7.022</td> <th>  Cond. No.          </th> <td>4.44e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.44e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.706\n",
       "Model:                            GLS   Adj. R-squared:                  0.684\n",
       "Method:                 Least Squares   F-statistic:                     32.37\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           6.74e-08\n",
       "Time:                        18:06:05   Log-Likelihood:                -192.19\n",
       "No. Observations:                  29   AIC:                             388.4\n",
       "Df Residuals:                      27   BIC:                             391.1\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households   654.3725    588.346      1.112      0.276    -552.813    1861.558\n",
       "ACCESS_Bike_LTS_Walk_JBS            0.0081      0.003      3.190      0.004       0.003       0.013\n",
       "==============================================================================\n",
       "Omnibus:                       13.323   Durbin-Watson:                   1.616\n",
       "Prob(Omnibus):                  0.001   Jarque-Bera (JB):               22.867\n",
       "Skew:                           0.829   Prob(JB):                     1.08e-05\n",
       "Kurtosis:                       7.022   Cond. No.                     4.44e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.44e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_mod = sm.GLS(reg_df[pm_ridership],reg_df[job_features])\n",
    "pm_res = pm_mod.fit()\n",
    "pm_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>AM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.800</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>       <th>  Adj. R-squared:    </th> <td>   0.786</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   54.11</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>3.58e-10</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -177.29</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   358.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>      <th>  BIC:               </th> <td>   361.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>  317.1798</td> <td>  351.996</td> <td>    0.901</td> <td> 0.376</td> <td> -405.056</td> <td> 1039.416</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>        <td>    0.0070</td> <td>    0.002</td> <td>    4.614</td> <td> 0.000</td> <td>    0.004</td> <td>    0.010</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>11.419</td> <th>  Durbin-Watson:     </th> <td>   2.148</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.003</td> <th>  Jarque-Bera (JB):  </th> <td>  10.927</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.111</td> <th>  Prob(JB):          </th> <td> 0.00424</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 5.026</td> <th>  Cond. No.          </th> <td>4.44e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.44e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         AM_Bike_Riders   R-squared:                       0.800\n",
       "Model:                            GLS   Adj. R-squared:                  0.786\n",
       "Method:                 Least Squares   F-statistic:                     54.11\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.58e-10\n",
       "Time:                        18:06:05   Log-Likelihood:                -177.29\n",
       "No. Observations:                  29   AIC:                             358.6\n",
       "Df Residuals:                      27   BIC:                             361.3\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households   317.1798    351.996      0.901      0.376    -405.056    1039.416\n",
       "ACCESS_Bike_LTS_Walk_JBS            0.0070      0.002      4.614      0.000       0.004       0.010\n",
       "==============================================================================\n",
       "Omnibus:                       11.419   Durbin-Watson:                   2.148\n",
       "Prob(Omnibus):                  0.003   Jarque-Bera (JB):               10.927\n",
       "Skew:                           1.111   Prob(JB):                      0.00424\n",
       "Kurtosis:                       5.026   Cond. No.                     4.44e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.44e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "am_mod = sm.GLS(reg_df[am_ridership],reg_df[job_features])\n",
    "am_res = am_mod.fit()\n",
    "am_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>GLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.706</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>GLS</td>       <th>  Adj. R-squared:    </th> <td>   0.684</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   32.37</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>6.74e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -192.19</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   388.4</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>      <th>  BIC:               </th> <td>   391.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>  654.3725</td> <td>  588.346</td> <td>    1.112</td> <td> 0.276</td> <td> -552.813</td> <td> 1861.558</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>        <td>    0.0081</td> <td>    0.003</td> <td>    3.190</td> <td> 0.004</td> <td>    0.003</td> <td>    0.013</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>13.323</td> <th>  Durbin-Watson:     </th> <td>   1.616</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.001</td> <th>  Jarque-Bera (JB):  </th> <td>  22.867</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.829</td> <th>  Prob(JB):          </th> <td>1.08e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 7.022</td> <th>  Cond. No.          </th> <td>4.44e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.44e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            GLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.706\n",
       "Model:                            GLS   Adj. R-squared:                  0.684\n",
       "Method:                 Least Squares   F-statistic:                     32.37\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           6.74e-08\n",
       "Time:                        18:06:05   Log-Likelihood:                -192.19\n",
       "No. Observations:                  29   AIC:                             388.4\n",
       "Df Residuals:                      27   BIC:                             391.1\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households   654.3725    588.346      1.112      0.276    -552.813    1861.558\n",
       "ACCESS_Bike_LTS_Walk_JBS            0.0081      0.003      3.190      0.004       0.003       0.013\n",
       "==============================================================================\n",
       "Omnibus:                       13.323   Durbin-Watson:                   1.616\n",
       "Prob(Omnibus):                  0.001   Jarque-Bera (JB):               22.867\n",
       "Skew:                           0.829   Prob(JB):                     1.08e-05\n",
       "Kurtosis:                       7.022   Cond. No.                     4.44e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.44e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_mod = sm.GLS(reg_df[pm_ridership],reg_df[job_features])\n",
    "pm_res = pm_mod.fit()\n",
    "pm_res.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the regression models and coefficient choices, there are very high R^2s with just the accessibility values (.5 to .8). Even though the significance values vary, models that produce reasonable coefficents and predictions with one control variable are commute to work via bicycle with job accessibility, and zero v"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.531</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>        <th>  Adj. R-squared:    </th> <td>   0.515</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   31.75</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>4.91e-06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -220.27</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   442.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    28</td>       <th>  BIC:               </th> <td>   443.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th> <td>    0.0210</td> <td>    0.004</td> <td>    5.635</td> <td> 0.000</td> <td>    0.013</td> <td>    0.029</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>32.974</td> <th>  Durbin-Watson:     </th> <td>   1.642</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  87.052</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 2.327</td> <th>  Prob(JB):          </th> <td>1.25e-19</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td>10.098</td> <th>  Cond. No.          </th> <td>    1.00</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.531\n",
       "Model:                            OLS   Adj. R-squared:                  0.515\n",
       "Method:                 Least Squares   F-statistic:                     31.75\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           4.91e-06\n",
       "Time:                        18:06:05   Log-Likelihood:                -220.27\n",
       "No. Observations:                  29   AIC:                             442.5\n",
       "Df Residuals:                      28   BIC:                             443.9\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "ACCESS_Bike_LTS_Walk_POP     0.0210      0.004      5.635      0.000       0.013       0.029\n",
       "==============================================================================\n",
       "Omnibus:                       32.974   Durbin-Watson:                   1.642\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               87.052\n",
       "Skew:                           2.327   Prob(JB):                     1.25e-19\n",
       "Kurtosis:                      10.098   Cond. No.                         1.00\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_pop_mod = sm.OLS(reg_df[daily_ridership],reg_df['ACCESS_Bike_LTS_Walk_POP'])\n",
    "daily_pop_res = daily_pop_mod.fit() \n",
    "daily_pop_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>AM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.794</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.787</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   108.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>4.03e-11</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -177.72</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   357.4</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    28</td>      <th>  BIC:               </th> <td>   358.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th> <td>    0.0082</td> <td>    0.001</td> <td>   10.399</td> <td> 0.000</td> <td>    0.007</td> <td>    0.010</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 7.992</td> <th>  Durbin-Watson:     </th> <td>   2.219</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.018</td> <th>  Jarque-Bera (JB):  </th> <td>   7.238</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.735</td> <th>  Prob(JB):          </th> <td>  0.0268</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.957</td> <th>  Cond. No.          </th> <td>    1.00</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         AM_Bike_Riders   R-squared:                       0.794\n",
       "Model:                            OLS   Adj. R-squared:                  0.787\n",
       "Method:                 Least Squares   F-statistic:                     108.1\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           4.03e-11\n",
       "Time:                        18:06:05   Log-Likelihood:                -177.72\n",
       "No. Observations:                  29   AIC:                             357.4\n",
       "Df Residuals:                      28   BIC:                             358.8\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "ACCESS_Bike_LTS_Walk_JBS     0.0082      0.001     10.399      0.000       0.007       0.010\n",
       "==============================================================================\n",
       "Omnibus:                        7.992   Durbin-Watson:                   2.219\n",
       "Prob(Omnibus):                  0.018   Jarque-Bera (JB):                7.238\n",
       "Skew:                           0.735   Prob(JB):                       0.0268\n",
       "Kurtosis:                       4.957   Cond. No.                         1.00\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "am_job_mod = sm.OLS(reg_df[am_ridership],reg_df['ACCESS_Bike_LTS_Walk_JBS'])\n",
    "am_job_res = am_job_mod.fit() \n",
    "am_job_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>AM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.662</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.650</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   54.83</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>4.60e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -184.93</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   371.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    28</td>      <th>  BIC:               </th> <td>   373.2</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th> <td>    0.0082</td> <td>    0.001</td> <td>    7.405</td> <td> 0.000</td> <td>    0.006</td> <td>    0.010</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>17.528</td> <th>  Durbin-Watson:     </th> <td>   1.925</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  21.425</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.530</td> <th>  Prob(JB):          </th> <td>2.23e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 5.892</td> <th>  Cond. No.          </th> <td>    1.00</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         AM_Bike_Riders   R-squared:                       0.662\n",
       "Model:                            OLS   Adj. R-squared:                  0.650\n",
       "Method:                 Least Squares   F-statistic:                     54.83\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           4.60e-08\n",
       "Time:                        18:06:05   Log-Likelihood:                -184.93\n",
       "No. Observations:                  29   AIC:                             371.9\n",
       "Df Residuals:                      28   BIC:                             373.2\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "ACCESS_Bike_LTS_Walk_POP     0.0082      0.001      7.405      0.000       0.006       0.010\n",
       "==============================================================================\n",
       "Omnibus:                       17.528   Durbin-Watson:                   1.925\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               21.425\n",
       "Skew:                           1.530   Prob(JB):                     2.23e-05\n",
       "Kurtosis:                       5.892   Cond. No.                         1.00\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "am_pop_mod = sm.OLS(reg_df[am_ridership],reg_df['ACCESS_Bike_LTS_Walk_POP'])\n",
    "am_pop_res = am_pop_mod.fit() \n",
    "am_pop_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.692</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.681</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   62.98</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>1.21e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -192.84</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   387.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    28</td>      <th>  BIC:               </th> <td>   389.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th> <td>    0.0106</td> <td>    0.001</td> <td>    7.936</td> <td> 0.000</td> <td>    0.008</td> <td>    0.013</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>13.038</td> <th>  Durbin-Watson:     </th> <td>   1.902</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.001</td> <th>  Jarque-Bera (JB):  </th> <td>  36.678</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.404</td> <th>  Prob(JB):          </th> <td>1.08e-08</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 8.450</td> <th>  Cond. No.          </th> <td>    1.00</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.692\n",
       "Model:                            OLS   Adj. R-squared:                  0.681\n",
       "Method:                 Least Squares   F-statistic:                     62.98\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           1.21e-08\n",
       "Time:                        18:06:05   Log-Likelihood:                -192.84\n",
       "No. Observations:                  29   AIC:                             387.7\n",
       "Df Residuals:                      28   BIC:                             389.0\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "ACCESS_Bike_LTS_Walk_JBS     0.0106      0.001      7.936      0.000       0.008       0.013\n",
       "==============================================================================\n",
       "Omnibus:                       13.038   Durbin-Watson:                   1.902\n",
       "Prob(Omnibus):                  0.001   Jarque-Bera (JB):               36.678\n",
       "Skew:                           0.404   Prob(JB):                     1.08e-08\n",
       "Kurtosis:                       8.450   Cond. No.                         1.00\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_job_mod = sm.OLS(reg_df[pm_ridership],reg_df['ACCESS_Bike_LTS_Walk_JBS'])\n",
    "pm_job_res = pm_job_mod.fit() \n",
    "pm_job_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.394</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.372</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   18.18</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>0.000207</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -202.67</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   407.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    28</td>      <th>  BIC:               </th> <td>   408.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th> <td>    0.0087</td> <td>    0.002</td> <td>    4.264</td> <td> 0.000</td> <td>    0.005</td> <td>    0.013</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>32.979</td> <th>  Durbin-Watson:     </th> <td>   1.521</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  80.025</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 2.410</td> <th>  Prob(JB):          </th> <td>4.20e-18</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 9.557</td> <th>  Cond. No.          </th> <td>    1.00</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.394\n",
       "Model:                            OLS   Adj. R-squared:                  0.372\n",
       "Method:                 Least Squares   F-statistic:                     18.18\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           0.000207\n",
       "Time:                        18:06:05   Log-Likelihood:                -202.67\n",
       "No. Observations:                  29   AIC:                             407.3\n",
       "Df Residuals:                      28   BIC:                             408.7\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "ACCESS_Bike_LTS_Walk_POP     0.0087      0.002      4.264      0.000       0.005       0.013\n",
       "==============================================================================\n",
       "Omnibus:                       32.979   Durbin-Watson:                   1.521\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               80.025\n",
       "Skew:                           2.410   Prob(JB):                     4.20e-18\n",
       "Kurtosis:                       9.557   Cond. No.                         1.00\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_pop_mod = sm.OLS(reg_df[pm_ridership],reg_df['ACCESS_Bike_LTS_Walk_POP'])\n",
    "pm_pop_res = pm_pop_mod.fit() \n",
    "pm_pop_res.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Interactive Regression Analysis\n",
    "Using Ipywidgets, different dependents/independents can be tested via a series of drop down menus. Helping to answer questions on the fly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ac49659d1a64c83a5899ed01400295e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='dependent_col', options=('Daily_Bike_Riders', 'AM_Bike_Riders', 'P…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.interactive_regression(reg_df, dependent_col, feature_list=[])>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def interactive_regression(reg_df, dependent_col,feature_list = []):\n",
    "    mod = sm.OLS(reg_df[dependent_col],reg_df[list(feature_list)])\n",
    "    result = mod.fit() \n",
    "    display(compute_elasticity(result,reg_df))\n",
    "    display(result.summary())\n",
    "widgets.interact(interactive_regression, reg_df=widgets.fixed(reg_df),dependent_col = [daily_ridership,am_ridership,pm_ridership], feature_list= \n",
    "                 widgets.SelectMultiple(options = all_independents,value= pop_features, disable=False,rows=5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chosen Variable Regression Model\n",
    "Based on exploring various predictive models, these are the regression results from the chosen model selection based the feature selection process above. The main findings that are used for the final model are the following:\n",
    "\n",
    "* The focus is on AM Bike Boarding as the focus is to inform the bike parking supply that should be at stations based on their relative typologies and low stress network connectivity\n",
    "\n",
    "* San Francisco is an outlier by enough that final models are developed with its exclusion. \n",
    "\n",
    "* Percent Zero Vehicles is a proxy for a multitude of behaviors and used as the control variables for these models. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.581</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>        <th>  Adj. R-squared:    </th> <td>   0.549</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   18.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>1.23e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -196.78</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>       <th>  AIC:               </th> <td>   397.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>       <th>  BIC:               </th> <td>   400.2</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td> 1656.2186</td> <td>  890.816</td> <td>    1.859</td> <td> 0.074</td> <td> -174.879</td> <td> 3487.316</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>        <td>    0.0105</td> <td>    0.005</td> <td>    2.211</td> <td> 0.036</td> <td>    0.001</td> <td>    0.020</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>10.572</td> <th>  Durbin-Watson:     </th> <td>   1.289</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.005</td> <th>  Jarque-Bera (JB):  </th> <td>   9.355</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.109</td> <th>  Prob(JB):          </th> <td> 0.00930</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.761</td> <th>  Cond. No.          </th> <td>2.93e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 2.93e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.581\n",
       "Model:                            OLS   Adj. R-squared:                  0.549\n",
       "Method:                 Least Squares   F-statistic:                     18.02\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           1.23e-05\n",
       "Time:                        18:06:05   Log-Likelihood:                -196.78\n",
       "No. Observations:                  28   AIC:                             397.6\n",
       "Df Residuals:                      26   BIC:                             400.2\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households  1656.2186    890.816      1.859      0.074    -174.879    3487.316\n",
       "ACCESS_Bike_LTS_Walk_JBS            0.0105      0.005      2.211      0.036       0.001       0.020\n",
       "==============================================================================\n",
       "Omnibus:                       10.572   Durbin-Watson:                   1.289\n",
       "Prob(Omnibus):                  0.005   Jarque-Bera (JB):                9.355\n",
       "Skew:                           1.109   Prob(JB):                      0.00930\n",
       "Kurtosis:                       4.761   Cond. No.                     2.93e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 2.93e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_final_job_mod = sm.OLS(reg_df[daily_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[job_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "daily_final_job_res = daily_final_job_mod.fit() \n",
    "daily_final_job_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.875</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>        <th>  Adj. R-squared:    </th> <td>   0.866</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   94.76</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>6.23e-13</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -201.07</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>       <th>  AIC:               </th> <td>   406.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>       <th>  BIC:               </th> <td>   408.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Jobs Within 1/2 Mile</th>     <td>    0.0439</td> <td>    0.011</td> <td>    4.182</td> <td> 0.000</td> <td>    0.022</td> <td>    0.065</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th> <td>    0.0093</td> <td>    0.004</td> <td>    2.422</td> <td> 0.022</td> <td>    0.001</td> <td>    0.017</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 6.808</td> <th>  Durbin-Watson:     </th> <td>   1.859</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.033</td> <th>  Jarque-Bera (JB):  </th> <td>   5.492</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.683</td> <th>  Prob(JB):          </th> <td>  0.0642</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.636</td> <th>  Cond. No.          </th> <td>    6.47</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.875\n",
       "Model:                            OLS   Adj. R-squared:                  0.866\n",
       "Method:                 Least Squares   F-statistic:                     94.76\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           6.23e-13\n",
       "Time:                        18:06:05   Log-Likelihood:                -201.07\n",
       "No. Observations:                  29   AIC:                             406.1\n",
       "Df Residuals:                      27   BIC:                             408.9\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "Jobs Within 1/2 Mile         0.0439      0.011      4.182      0.000       0.022       0.065\n",
       "ACCESS_Bike_LTS_Walk_JBS     0.0093      0.004      2.422      0.022       0.001       0.017\n",
       "==============================================================================\n",
       "Omnibus:                        6.808   Durbin-Watson:                   1.859\n",
       "Prob(Omnibus):                  0.033   Jarque-Bera (JB):                5.492\n",
       "Skew:                           0.683   Prob(JB):                       0.0642\n",
       "Kurtosis:                       4.636   Cond. No.                         6.47\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_final_job_wjobcontrol_mod = sm.OLS(reg_df[daily_ridership],reg_df[job_control_features])\n",
    "daily_final_job_wjobcontrol_res = daily_final_job_wjobcontrol_mod.fit() \n",
    "daily_final_job_wjobcontrol_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.813</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.800</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   58.85</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>1.44e-10</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -185.58</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    29</td>      <th>  AIC:               </th> <td>   375.2</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    27</td>      <th>  BIC:               </th> <td>   377.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Jobs Within 1/2 Mile</th>     <td>    0.0258</td> <td>    0.006</td> <td>    4.187</td> <td> 0.000</td> <td>    0.013</td> <td>    0.038</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th> <td>    0.0022</td> <td>    0.002</td> <td>    0.967</td> <td> 0.342</td> <td>   -0.002</td> <td>    0.007</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>16.281</td> <th>  Durbin-Watson:     </th> <td>   1.335</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  20.089</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.380</td> <th>  Prob(JB):          </th> <td>4.34e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 6.002</td> <th>  Cond. No.          </th> <td>    6.47</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.813\n",
       "Model:                            OLS   Adj. R-squared:                  0.800\n",
       "Method:                 Least Squares   F-statistic:                     58.85\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           1.44e-10\n",
       "Time:                        18:06:05   Log-Likelihood:                -185.58\n",
       "No. Observations:                  29   AIC:                             375.2\n",
       "Df Residuals:                      27   BIC:                             377.9\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "============================================================================================\n",
       "                               coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------------\n",
       "Jobs Within 1/2 Mile         0.0258      0.006      4.187      0.000       0.013       0.038\n",
       "ACCESS_Bike_LTS_Walk_JBS     0.0022      0.002      0.967      0.342      -0.002       0.007\n",
       "==============================================================================\n",
       "Omnibus:                       16.281   Durbin-Watson:                   1.335\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               20.089\n",
       "Skew:                           1.380   Prob(JB):                     4.34e-05\n",
       "Kurtosis:                       6.002   Cond. No.                         6.47\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_final_job_wjobcontrol_mod = sm.OLS(reg_df[pm_ridership],reg_df[job_control_features])\n",
    "pm_final_job_wjobcontrol_res = pm_final_job_wjobcontrol_mod.fit() \n",
    "pm_final_job_wjobcontrol_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>AM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.611</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.581</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   20.39</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>4.73e-06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -171.46</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>      <th>  AIC:               </th> <td>   346.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>      <th>  BIC:               </th> <td>   349.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>  352.9375</td> <td>  360.663</td> <td>    0.979</td> <td> 0.337</td> <td> -388.416</td> <td> 1094.291</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>        <td>    0.0063</td> <td>    0.002</td> <td>    3.272</td> <td> 0.003</td> <td>    0.002</td> <td>    0.010</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>13.453</td> <th>  Durbin-Watson:     </th> <td>   1.956</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.001</td> <th>  Jarque-Bera (JB):  </th> <td>  13.346</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.331</td> <th>  Prob(JB):          </th> <td> 0.00126</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 5.087</td> <th>  Cond. No.          </th> <td>2.93e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 2.93e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         AM_Bike_Riders   R-squared:                       0.611\n",
       "Model:                            OLS   Adj. R-squared:                  0.581\n",
       "Method:                 Least Squares   F-statistic:                     20.39\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           4.73e-06\n",
       "Time:                        18:06:05   Log-Likelihood:                -171.46\n",
       "No. Observations:                  28   AIC:                             346.9\n",
       "Df Residuals:                      26   BIC:                             349.6\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households   352.9375    360.663      0.979      0.337    -388.416    1094.291\n",
       "ACCESS_Bike_LTS_Walk_JBS            0.0063      0.002      3.272      0.003       0.002       0.010\n",
       "==============================================================================\n",
       "Omnibus:                       13.453   Durbin-Watson:                   1.956\n",
       "Prob(Omnibus):                  0.001   Jarque-Bera (JB):               13.346\n",
       "Skew:                           1.331   Prob(JB):                      0.00126\n",
       "Kurtosis:                       5.087   Cond. No.                     2.93e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 2.93e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "am_final_job_mod = sm.OLS(reg_df[am_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[job_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "am_final_job_res = am_final_job_mod.fit() \n",
    "am_final_job_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.400</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.354</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   8.679</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th>  <td>0.00130</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -181.39</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>      <th>  AIC:               </th> <td>   366.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>      <th>  BIC:               </th> <td>   369.4</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>  917.0027</td> <td>  514.063</td> <td>    1.784</td> <td> 0.086</td> <td> -139.670</td> <td> 1973.675</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_JBS</th>        <td>    0.0028</td> <td>    0.003</td> <td>    1.029</td> <td> 0.313</td> <td>   -0.003</td> <td>    0.008</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>31.864</td> <th>  Durbin-Watson:     </th> <td>   0.971</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  78.239</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 2.309</td> <th>  Prob(JB):          </th> <td>1.02e-17</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 9.763</td> <th>  Cond. No.          </th> <td>2.93e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 2.93e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.400\n",
       "Model:                            OLS   Adj. R-squared:                  0.354\n",
       "Method:                 Least Squares   F-statistic:                     8.679\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):            0.00130\n",
       "Time:                        18:06:05   Log-Likelihood:                -181.39\n",
       "No. Observations:                  28   AIC:                             366.8\n",
       "Df Residuals:                      26   BIC:                             369.4\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households   917.0027    514.063      1.784      0.086    -139.670    1973.675\n",
       "ACCESS_Bike_LTS_Walk_JBS            0.0028      0.003      1.029      0.313      -0.003       0.008\n",
       "==============================================================================\n",
       "Omnibus:                       31.864   Durbin-Watson:                   0.971\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               78.239\n",
       "Skew:                           2.309   Prob(JB):                     1.02e-17\n",
       "Kurtosis:                       9.763   Cond. No.                     2.93e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 2.93e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_final_job_mod = sm.OLS(reg_df[pm_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[job_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "pm_final_job_res = pm_final_job_mod.fit() \n",
    "pm_final_job_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>Daily_Bike_Riders</td> <th>  R-squared:         </th> <td>   0.542</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>        <th>  Adj. R-squared:    </th> <td>   0.507</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>   <th>  F-statistic:       </th> <td>   15.41</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td>  <th>  Prob (F-statistic):</th> <td>3.85e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>      <th>  Log-Likelihood:    </th> <td> -198.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>       <th>  AIC:               </th> <td>   400.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>       <th>  BIC:               </th> <td>   402.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>       <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>     <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td> 1783.2444</td> <td> 1093.691</td> <td>    1.630</td> <td> 0.115</td> <td> -464.870</td> <td> 4031.358</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>    0.0070</td> <td>    0.005</td> <td>    1.513</td> <td> 0.142</td> <td>   -0.003</td> <td>    0.017</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>15.482</td> <th>  Durbin-Watson:     </th> <td>   1.211</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  16.496</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.488</td> <th>  Prob(JB):          </th> <td>0.000262</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 5.297</td> <th>  Cond. No.          </th> <td>4.35e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.35e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:      Daily_Bike_Riders   R-squared:                       0.542\n",
       "Model:                            OLS   Adj. R-squared:                  0.507\n",
       "Method:                 Least Squares   F-statistic:                     15.41\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.85e-05\n",
       "Time:                        18:06:05   Log-Likelihood:                -198.01\n",
       "No. Observations:                  28   AIC:                             400.0\n",
       "Df Residuals:                      26   BIC:                             402.7\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households  1783.2444   1093.691      1.630      0.115    -464.870    4031.358\n",
       "ACCESS_Bike_LTS_Walk_POP            0.0070      0.005      1.513      0.142      -0.003       0.017\n",
       "==============================================================================\n",
       "Omnibus:                       15.482   Durbin-Watson:                   1.211\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               16.496\n",
       "Skew:                           1.488   Prob(JB):                     0.000262\n",
       "Kurtosis:                       5.297   Cond. No.                     4.35e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.35e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_final_pop_mod = sm.OLS(reg_df[daily_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[pop_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "daily_final_pop_res = daily_final_pop_mod.fit() \n",
    "daily_final_pop_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>AM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.622</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.593</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   21.42</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th> <td>3.19e-06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -171.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>      <th>  AIC:               </th> <td>   346.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>      <th>  BIC:               </th> <td>   348.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td>   55.7502</td> <td>  417.386</td> <td>    0.134</td> <td> 0.895</td> <td> -802.199</td> <td>  913.699</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>    0.0061</td> <td>    0.002</td> <td>    3.441</td> <td> 0.002</td> <td>    0.002</td> <td>    0.010</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>13.516</td> <th>  Durbin-Watson:     </th> <td>   2.038</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.001</td> <th>  Jarque-Bera (JB):  </th> <td>  13.363</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 1.347</td> <th>  Prob(JB):          </th> <td> 0.00125</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 5.049</td> <th>  Cond. No.          </th> <td>4.35e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.35e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         AM_Bike_Riders   R-squared:                       0.622\n",
       "Model:                            OLS   Adj. R-squared:                  0.593\n",
       "Method:                 Least Squares   F-statistic:                     21.42\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):           3.19e-06\n",
       "Time:                        18:06:05   Log-Likelihood:                -171.04\n",
       "No. Observations:                  28   AIC:                             346.1\n",
       "Df Residuals:                      26   BIC:                             348.7\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households    55.7502    417.386      0.134      0.895    -802.199     913.699\n",
       "ACCESS_Bike_LTS_Walk_POP            0.0061      0.002      3.441      0.002       0.002       0.010\n",
       "==============================================================================\n",
       "Omnibus:                       13.516   Durbin-Watson:                   2.038\n",
       "Prob(Omnibus):                  0.001   Jarque-Bera (JB):               13.363\n",
       "Skew:                           1.347   Prob(JB):                      0.00125\n",
       "Kurtosis:                       5.049   Cond. No.                     4.35e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.35e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "am_final_pop_mod = sm.OLS(reg_df[am_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[pop_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "am_final_pop_res = am_final_pop_mod.fit() \n",
    "am_final_pop_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>PM_Bike_Riders</td>  <th>  R-squared:         </th> <td>   0.376</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.328</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   7.833</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Tue, 05 Feb 2019</td> <th>  Prob (F-statistic):</th>  <td>0.00217</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>18:06:05</td>     <th>  Log-Likelihood:    </th> <td> -181.94</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    28</td>      <th>  AIC:               </th> <td>   367.9</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    26</td>      <th>  BIC:               </th> <td>   370.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Percent Zero Vehicle Households</th> <td> 1348.8555</td> <td>  616.150</td> <td>    2.189</td> <td> 0.038</td> <td>   82.341</td> <td> 2615.370</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ACCESS_Bike_LTS_Walk_POP</th>        <td>   -0.0001</td> <td>    0.003</td> <td>   -0.048</td> <td> 0.962</td> <td>   -0.005</td> <td>    0.005</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>28.279</td> <th>  Durbin-Watson:     </th> <td>   0.931</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  56.578</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 2.144</td> <th>  Prob(JB):          </th> <td>5.18e-13</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 8.487</td> <th>  Cond. No.          </th> <td>4.35e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 4.35e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         PM_Bike_Riders   R-squared:                       0.376\n",
       "Model:                            OLS   Adj. R-squared:                  0.328\n",
       "Method:                 Least Squares   F-statistic:                     7.833\n",
       "Date:                Tue, 05 Feb 2019   Prob (F-statistic):            0.00217\n",
       "Time:                        18:06:05   Log-Likelihood:                -181.94\n",
       "No. Observations:                  28   AIC:                             367.9\n",
       "Df Residuals:                      26   BIC:                             370.6\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Percent Zero Vehicle Households  1348.8555    616.150      2.189      0.038      82.341    2615.370\n",
       "ACCESS_Bike_LTS_Walk_POP           -0.0001      0.003     -0.048      0.962      -0.005       0.005\n",
       "==============================================================================\n",
       "Omnibus:                       28.279   Durbin-Watson:                   0.931\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               56.578\n",
       "Skew:                           2.144   Prob(JB):                     5.18e-13\n",
       "Kurtosis:                       8.487   Cond. No.                     4.35e+05\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 4.35e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm_final_pop_mod = sm.OLS(reg_df[pm_ridership][reg_df[\"Name\"]!=\"4th and King\"],reg_df[pop_features][reg_df[\"Name\"]!=\"4th and King\"])\n",
    "pm_final_pop_res = pm_final_pop_mod.fit() \n",
    "pm_final_pop_res.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Derive Elasticity\n",
    "Based on the chosen model, elasticities are computed for job and population accessibiltiy via low stress networks. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Low Stress Job Accessibility Elasticities\n",
    "Derive elasticities of distance decay low stress bicycling job accessibility. Impedances assume a user walks on high stress segments, and bikes at normal speeds on low stress segments. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 86.6%, transparent 86.6%);\n",
       "        }    #T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col2\" class=\"data row0 col2\" >1656.22</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.0743498</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.415106</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.0105435</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.0360695</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.479409</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col1\" class=\"data row2 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col3\" class=\"data row2 col3\" >374.069</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c2feb4a2_29b3_11e9_ac67_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x21306910780>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_final_job_res,reg_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 56.0%, transparent 56.0%);\n",
       "        }    #T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col2\" class=\"data row0 col2\" >917.003</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.0861284</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.578743</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.0028314</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.313095</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.324188</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col1\" class=\"data row2 col1\" >PM_Bike_Riders</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col3\" class=\"data row2 col3\" >148.552</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c3031e74_29b3_11e9_81df_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2130b83eeb8>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(pm_final_job_res,reg_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 30.8%, transparent 30.8%);\n",
       "        }    #T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col2\" class=\"data row0 col2\" >352.937</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.336806</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.215785</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.0063191</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.00300953</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.700905</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col1\" class=\"data row2 col1\" >AM_Bike_Riders</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col3\" class=\"data row2 col3\" >153.345</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c307137a_29b3_11e9_bff3_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x21306b11f60>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(am_final_job_res,reg_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Controlling for Jobs Density vs. Low Stress Job Access\n",
    "Accessibility as a metric INTRINSICALLY connects how a network functions (range or coverage possible across modes) with the land use patterns of a study area (density, clustering of destinations/origins). While we can say the elasticities of \"low stress job accessibility\" using soft impedances is in the range .8 to .93, but how much of this is related to built density? This elasticity check compares the contributions of level of traffic stress to daily ridership and pm ridership when looking at jobs within a half mile. Jobs within a half mile maintains significance, and so does low stress accessibility when compared to daily. \n",
    "\n",
    "When you evaluate jobs within a half mile vs. low stress job accessiblity, low stress accessibility maintains some impact as a contributor to daily ridership (even though the model has a high degree of collinearity relative as they measure in essense the same thing- jobs). That said the elasticity is significantly lower for low stress job accessiblity when compared to jobs within a half mile. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 51.1%, transparent 51.1%);\n",
       "        }    #T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col1\" class=\"data row0 col1\" >Jobs Within 1/2 Mile</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col2\" class=\"data row0 col2\" >0.0439387</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col3\" class=\"data row0 col3\" >7079.99</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.000273099</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.831626</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.00934216</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.0224327</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.424784</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col1\" class=\"data row2 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col3\" class=\"data row2 col3\" >374.069</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c30b56ae_29b3_11e9_8cab_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2130a5a5ba8>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_final_job_wjobcontrol_res,reg_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 20.4%, transparent 20.4%);\n",
       "        }    #T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col1\" class=\"data row0 col1\" >Jobs Within 1/2 Mile</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col2\" class=\"data row0 col2\" >0.0257886</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col3\" class=\"data row0 col3\" >7079.99</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.000269051</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row0_col5\" class=\"data row0 col5\" >1.22909</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.00218595</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.342221</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.250286</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col1\" class=\"data row2 col1\" >PM_Bike_Riders</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col3\" class=\"data row2 col3\" >148.552</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c30f990c_29b3_11e9_b770_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2130b801668>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(pm_final_job_wjobcontrol_res,reg_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Removing San Francisco\n",
    "San Francisco has an outside impact on the results given its density and size. When removed, we still see positive elasticity, but it reduces. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 86.6%, transparent 86.6%);\n",
       "        }    #T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col2\" class=\"data row0 col2\" >1656.22</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.0743498</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.415106</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_JBS</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.0105435</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col3\" class=\"data row1 col3\" >17008.8</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.0360695</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.479409</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col1\" class=\"data row2 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col3\" class=\"data row2 col3\" >374.069</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c3138de4_29b3_11e9_9378_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x213068c1240>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_final_job_res_no_San_Francisco,reg_df[reg_df[\"Name\"]!=\"4th & King\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 88.8%, transparent 88.8%);\n",
       "        }    #T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col2\" class=\"data row0 col2\" >1783.24</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.115056</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.446943</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_POP</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.00700239</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col3\" class=\"data row1 col3\" >21192.6</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.142408</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.396715</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col1\" class=\"data row2 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col3\" class=\"data row2 col3\" >374.069</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c317d7a4_29b3_11e9_a855_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x213068c1780>"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_final_pop_res_no_San_Francisco,reg_df[reg_df[\"Name\"]!=\"4th & King\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Low Stress Population Accessibility Elasticities\n",
    "Derive elasticities of distance decay low stress bicycling population accessibility. Impedances assume a user walks on high stress segments, and bikes at normal speeds on low stress segments. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 88.8%, transparent 88.8%);\n",
       "        }    #T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col2\" class=\"data row0 col2\" >1783.24</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.115056</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.446943</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_POP</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.00700239</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col3\" class=\"data row1 col3\" >21192.6</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.142408</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.396715</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col1\" class=\"data row2 col1\" >Daily_Bike_Riders</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col3\" class=\"data row2 col3\" >374.069</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c31b7ea8_29b3_11e9_9e5a_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x213068c1860>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(daily_final_pop_res,reg_df[reg_df[\"Name\"]!=\"4th & King\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 2.0%, #5fba7d 2.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #d65f5f 0.0%, #d65f5f 2.0%, transparent 2.0%);\n",
       "        }    #T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 2.0%, transparent 2.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col2\" class=\"data row0 col2\" >1348.86</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.0377571</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.851295</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_POP</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col2\" class=\"data row1 col2\" >-0.000124256</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col3\" class=\"data row1 col3\" >21192.6</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.962361</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row1_col5\" class=\"data row1 col5\" >-0.0177265</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col1\" class=\"data row2 col1\" >PM_Bike_Riders</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col3\" class=\"data row2 col3\" >148.552</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c31f738c_29b3_11e9_aec3_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x213068c11d0>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(pm_final_pop_res,reg_df[reg_df[\"Name\"]!=\"4th & King\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 4.1%, transparent 4.1%);\n",
       "        }    #T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent 0.0%, #5fba7d 0.0%, #5fba7d 100.0%, transparent 100.0%);\n",
       "        }    #T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col5 {\n",
       "            width:  10em;\n",
       "             height:  80%;\n",
       "            background:  linear-gradient(90deg, transparent 0%, transparent nan%, #d65f5f nan%, #d65f5f 0.0%, transparent 0.0%);\n",
       "        }</style>  \n",
       "<table id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" >Variable Type</th> \n",
       "        <th class=\"col_heading level0 col1\" >Variable</th> \n",
       "        <th class=\"col_heading level0 col2\" >Coefficient</th> \n",
       "        <th class=\"col_heading level0 col3\" >Variable Mean</th> \n",
       "        <th class=\"col_heading level0 col4\" >Significance</th> \n",
       "        <th class=\"col_heading level0 col5\" >Elasticity</th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col0\" class=\"data row0 col0\" >Independent</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col1\" class=\"data row0 col1\" >Percent Zero Vehicle Households</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col2\" class=\"data row0 col2\" >55.7502</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col3\" class=\"data row0 col3\" >0.0937546</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col4\" class=\"data row0 col4\" >0.894771</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row0_col5\" class=\"data row0 col5\" >0.0340855</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col0\" class=\"data row1 col0\" >Independent</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col1\" class=\"data row1 col1\" >ACCESS_Bike_LTS_Walk_POP</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col2\" class=\"data row1 col2\" >0.0060784</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col3\" class=\"data row1 col3\" >21192.6</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col4\" class=\"data row1 col4\" >0.00197073</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row1_col5\" class=\"data row1 col5\" >0.840048</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col0\" class=\"data row2 col0\" >Dependent</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col1\" class=\"data row2 col1\" >AM_Bike_Riders</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col2\" class=\"data row2 col2\" >nan</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col3\" class=\"data row2 col3\" >153.345</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col4\" class=\"data row2 col4\" >nan</td> \n",
       "        <td id=\"T_c322f39e_29b3_11e9_9774_10f005ecad14row2_col5\" class=\"data row2 col5\" >nan</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x213068ad198>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compute_elasticity(am_final_pop_res,reg_df[reg_df[\"Name\"]!=\"4th & King\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results Discussion\n",
    "\n",
    "* Low Stress Job Accessibility Maintains a consistent significant predictor of bicycle boardings when considering daily, pm, and am bicycle boardings. Elasticities in the range of .8 to .93 keep the elasticities predicted within the ranges of other regression models even with other/more control variables. This means we can expect about a 1% increase in low stress job accessibility we expect a .8 to .93% increase bicycle boardings at Caltrain Stations. \n",
    "\n",
    " * Low stress job accessibility measures the contribution of the network (level of traffic stress/network connectivity) and its intersection of land use (density/jobs) together in one metric. When you control for jobs, the contribution from just LTS alone reduces to about an elasticity of .4 of daily ridership. This is to say that LTS is potentially making a contribution to predicting boardings, but you have to keep in mind that low stress accessiblity is metric of land use & network characteeristics. \n",
    "\n",
    " * When you remove San Francisco from the analysis, the R^2s of the model can change significantly. When removed, elasticities get closer to 0.5.\n",
    "\n",
    "* Low Stress Population Accessibility has an unstable relationship with bicycle boardings. It has relatively strong elasticities for daily and pm bicycle boardings, BUT none of those results are significant (p-value < .05). In the case of AM Boardings, it does have a stronger relationship and elasticity relative job accessibility. Since this elasticity is within the same order of magnitude of job accessibility elastcities, more significant in the AM (more relevant time period), and signfiicant when high density outliers that might not apply to ST Express- it is the chosen elasticity we have decided to use for our analysis. "
   ]
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}