3. Synthetic Control Method.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1cdabe8b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:58:33.428955Z",
     "start_time": "2022-04-14T22:58:33.418368Z"
    }
   },
   "outputs": [],
   "source": [
    "# import pandas as pd \n",
    "\n",
    "# path = 'raw/covidinfo/utla_2022-02-25.csv'\n",
    "# df = pd.read_csv(path)\n",
    "# df['date'] = pd.to_datetime(df['date'])\n",
    "# df = df.sort_values(['areaCode', 'date']).reset_index(drop = True)\n",
    "# df\n",
    "\n",
    "# df['areaCode'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcead71f",
   "metadata": {},
   "source": [
    "# ltla Covariates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f1a9e575",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:23.748828Z",
     "start_time": "2022-05-10T04:12:23.188904Z"
    }
   },
   "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>areaCode</th>\n",
       "      <th>areaName</th>\n",
       "      <th>NHSER21CD</th>\n",
       "      <th>NHSER21NM</th>\n",
       "      <th>old_dependency_ratio</th>\n",
       "      <th>young_dependency_ratio</th>\n",
       "      <th>popu_all</th>\n",
       "      <th>Median age</th>\n",
       "      <th>People per sq. km</th>\n",
       "      <th>Average Score</th>\n",
       "      <th>Area (sq km)</th>\n",
       "      <th>male_avg_age</th>\n",
       "      <th>female_avg_age</th>\n",
       "      <th>female_rate</th>\n",
       "      <th>areaidx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>E40000009</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.288634</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>E06000002</td>\n",
       "      <td>Middlesbrough</td>\n",
       "      <td>E40000009</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>0.258597</td>\n",
       "      <td>0.315276</td>\n",
       "      <td>141285</td>\n",
       "      <td>36.3</td>\n",
       "      <td>2616</td>\n",
       "      <td>40.5</td>\n",
       "      <td>53.8816</td>\n",
       "      <td>37.015974</td>\n",
       "      <td>39.302367</td>\n",
       "      <td>0.505057</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>E06000003</td>\n",
       "      <td>Redcar and Cleveland</td>\n",
       "      <td>E40000009</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>0.378299</td>\n",
       "      <td>0.280907</td>\n",
       "      <td>137228</td>\n",
       "      <td>45.2</td>\n",
       "      <td>560</td>\n",
       "      <td>29.8</td>\n",
       "      <td>245.0698</td>\n",
       "      <td>42.037639</td>\n",
       "      <td>44.115800</td>\n",
       "      <td>0.514633</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>E06000004</td>\n",
       "      <td>Stockton-on-Tees</td>\n",
       "      <td>E40000009</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>0.299317</td>\n",
       "      <td>0.299981</td>\n",
       "      <td>197419</td>\n",
       "      <td>40.7</td>\n",
       "      <td>963</td>\n",
       "      <td>25.8</td>\n",
       "      <td>204.9331</td>\n",
       "      <td>39.646713</td>\n",
       "      <td>41.726862</td>\n",
       "      <td>0.505058</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>E06000005</td>\n",
       "      <td>Darlington</td>\n",
       "      <td>E40000009</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>0.333318</td>\n",
       "      <td>0.284279</td>\n",
       "      <td>107402</td>\n",
       "      <td>43.2</td>\n",
       "      <td>541</td>\n",
       "      <td>25.7</td>\n",
       "      <td>197.4758</td>\n",
       "      <td>41.155788</td>\n",
       "      <td>42.778638</td>\n",
       "      <td>0.513445</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>300</th>\n",
       "      <td>E09000029</td>\n",
       "      <td>Sutton</td>\n",
       "      <td>E40000003</td>\n",
       "      <td>London</td>\n",
       "      <td>0.238610</td>\n",
       "      <td>0.314247</td>\n",
       "      <td>207707</td>\n",
       "      <td>39.5</td>\n",
       "      <td>4706</td>\n",
       "      <td>14.0</td>\n",
       "      <td>43.8479</td>\n",
       "      <td>38.056436</td>\n",
       "      <td>39.842322</td>\n",
       "      <td>0.512202</td>\n",
       "      <td>301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>301</th>\n",
       "      <td>E09000030</td>\n",
       "      <td>Tower Hamlets</td>\n",
       "      <td>E40000003</td>\n",
       "      <td>London</td>\n",
       "      <td>0.088241</td>\n",
       "      <td>0.257283</td>\n",
       "      <td>331969</td>\n",
       "      <td>31.6</td>\n",
       "      <td>16427</td>\n",
       "      <td>27.9</td>\n",
       "      <td>19.7710</td>\n",
       "      <td>32.614445</td>\n",
       "      <td>32.569271</td>\n",
       "      <td>0.474219</td>\n",
       "      <td>302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>302</th>\n",
       "      <td>E09000031</td>\n",
       "      <td>Waltham Forest</td>\n",
       "      <td>E40000003</td>\n",
       "      <td>London</td>\n",
       "      <td>0.161879</td>\n",
       "      <td>0.307642</td>\n",
       "      <td>276940</td>\n",
       "      <td>35.2</td>\n",
       "      <td>7137</td>\n",
       "      <td>25.2</td>\n",
       "      <td>38.8102</td>\n",
       "      <td>35.037225</td>\n",
       "      <td>36.734122</td>\n",
       "      <td>0.495299</td>\n",
       "      <td>303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303</th>\n",
       "      <td>E09000032</td>\n",
       "      <td>Wandsworth</td>\n",
       "      <td>E40000003</td>\n",
       "      <td>London</td>\n",
       "      <td>0.133132</td>\n",
       "      <td>0.238689</td>\n",
       "      <td>329735</td>\n",
       "      <td>34.0</td>\n",
       "      <td>9624</td>\n",
       "      <td>16.6</td>\n",
       "      <td>34.2619</td>\n",
       "      <td>34.995129</td>\n",
       "      <td>35.965513</td>\n",
       "      <td>0.520603</td>\n",
       "      <td>304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>304</th>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>E40000003</td>\n",
       "      <td>London</td>\n",
       "      <td>0.175793</td>\n",
       "      <td>0.223543</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "      <td>305</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>305 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      areaCode              areaName  NHSER21CD                 NHSER21NM  \\\n",
       "0    E06000001            Hartlepool  E40000009  North East and Yorkshire   \n",
       "1    E06000002         Middlesbrough  E40000009  North East and Yorkshire   \n",
       "2    E06000003  Redcar and Cleveland  E40000009  North East and Yorkshire   \n",
       "3    E06000004      Stockton-on-Tees  E40000009  North East and Yorkshire   \n",
       "4    E06000005            Darlington  E40000009  North East and Yorkshire   \n",
       "..         ...                   ...        ...                       ...   \n",
       "300  E09000029                Sutton  E40000003                    London   \n",
       "301  E09000030         Tower Hamlets  E40000003                    London   \n",
       "302  E09000031        Waltham Forest  E40000003                    London   \n",
       "303  E09000032            Wandsworth  E40000003                    London   \n",
       "304  E09000033           Westminster  E40000003                    London   \n",
       "\n",
       "     old_dependency_ratio  young_dependency_ratio  popu_all  Median age  \\\n",
       "0                0.313702                0.288634     93836        41.7   \n",
       "1                0.258597                0.315276    141285        36.3   \n",
       "2                0.378299                0.280907    137228        45.2   \n",
       "3                0.299317                0.299981    197419        40.7   \n",
       "4                0.333318                0.284279    107402        43.2   \n",
       "..                    ...                     ...       ...         ...   \n",
       "300              0.238610                0.314247    207707        39.5   \n",
       "301              0.088241                0.257283    331969        31.6   \n",
       "302              0.161879                0.307642    276940        35.2   \n",
       "303              0.133132                0.238689    329735        34.0   \n",
       "304              0.175793                0.223543    269848        36.3   \n",
       "\n",
       "     People per sq. km  Average Score  Area (sq km)  male_avg_age  \\\n",
       "0                  999           35.0       93.7125     40.157418   \n",
       "1                 2616           40.5       53.8816     37.015974   \n",
       "2                  560           29.8      245.0698     42.037639   \n",
       "3                  963           25.8      204.9331     39.646713   \n",
       "4                  541           25.7      197.4758     41.155788   \n",
       "..                 ...            ...           ...           ...   \n",
       "300               4706           14.0       43.8479     38.056436   \n",
       "301              16427           27.9       19.7710     32.614445   \n",
       "302               7137           25.2       38.8102     35.037225   \n",
       "303               9624           16.6       34.2619     34.995129   \n",
       "304              12167           20.3       21.4826     36.918846   \n",
       "\n",
       "     female_avg_age  female_rate  areaidx  \n",
       "0         42.190274     0.509730        1  \n",
       "1         39.302367     0.505057        2  \n",
       "2         44.115800     0.514633        3  \n",
       "3         41.726862     0.505058        4  \n",
       "4         42.778638     0.513445        5  \n",
       "..              ...          ...      ...  \n",
       "300       39.842322     0.512202      301  \n",
       "301       32.569271     0.474219      302  \n",
       "302       36.734122     0.495299      303  \n",
       "303       35.965513     0.520603      304  \n",
       "304       38.427939     0.472677      305  \n",
       "\n",
       "[305 rows x 15 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "path = 'data/ltla_neighbours_df.csv'\n",
    "ltla_neighbours_df = pd.read_csv(path)# .rename(columns = {'LAD21NM': 'areaName',})\n",
    "ltla_neighbours_df['neighboors'] = ltla_neighbours_df['neighboors'].apply(eval)\n",
    "# ltla_neighbours_df\n",
    "\n",
    "path = 'data/ltla_region_england.csv'\n",
    "ltla_region_england_df = pd.read_csv(path).rename(columns = {'LAD21CD': 'areaCode',\n",
    "                                                            'LAD21NM': 'areaName',})\n",
    "path = 'data/ltal_covariates_df.csv'\n",
    "ltla_covariates_df = pd.read_csv(path)\n",
    "ltla_df = pd.merge(ltla_region_england_df, ltla_covariates_df)\n",
    "# ltla_df = pd.merge(ltla_df, ltla_neighbours_df)\n",
    "ltla_df['areaidx'] = list(range(1, len(ltla_df) + 1))\n",
    "\n",
    "ltla_df.columns\n",
    "ltla_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6bfca84",
   "metadata": {},
   "source": [
    "# ltla covid info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "aeec839a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.068235Z",
     "start_time": "2022-05-10T04:12:25.045566Z"
    }
   },
   "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>areaCode</th>\n",
       "      <th>areaName</th>\n",
       "      <th>areaType</th>\n",
       "      <th>date</th>\n",
       "      <th>cumCasesBySpecimenDate</th>\n",
       "      <th>cumCasesBySpecimenDateRate</th>\n",
       "      <th>newCasesBySpecimenDate</th>\n",
       "      <th>dateidx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>1</td>\n",
       "      <td>1.1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2020-03-17</td>\n",
       "      <td>1</td>\n",
       "      <td>1.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2020-03-18</td>\n",
       "      <td>2</td>\n",
       "      <td>2.1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2020-03-19</td>\n",
       "      <td>2</td>\n",
       "      <td>2.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2020-03-20</td>\n",
       "      <td>3</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218886</th>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2022-02-20</td>\n",
       "      <td>58626</td>\n",
       "      <td>21725.6</td>\n",
       "      <td>137.0</td>\n",
       "      <td>781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218887</th>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2022-02-21</td>\n",
       "      <td>58767</td>\n",
       "      <td>21777.8</td>\n",
       "      <td>141.0</td>\n",
       "      <td>782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218888</th>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2022-02-22</td>\n",
       "      <td>58913</td>\n",
       "      <td>21831.9</td>\n",
       "      <td>146.0</td>\n",
       "      <td>783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218889</th>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2022-02-23</td>\n",
       "      <td>59035</td>\n",
       "      <td>21877.1</td>\n",
       "      <td>122.0</td>\n",
       "      <td>784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218890</th>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>2022-02-24</td>\n",
       "      <td>59072</td>\n",
       "      <td>21890.8</td>\n",
       "      <td>37.0</td>\n",
       "      <td>785</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>218891 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         areaCode     areaName areaType       date  cumCasesBySpecimenDate  \\\n",
       "0       E06000001   Hartlepool     ltla 2020-03-16                       1   \n",
       "1       E06000001   Hartlepool     ltla 2020-03-17                       1   \n",
       "2       E06000001   Hartlepool     ltla 2020-03-18                       2   \n",
       "3       E06000001   Hartlepool     ltla 2020-03-19                       2   \n",
       "4       E06000001   Hartlepool     ltla 2020-03-20                       3   \n",
       "...           ...          ...      ...        ...                     ...   \n",
       "218886  E09000033  Westminster     ltla 2022-02-20                   58626   \n",
       "218887  E09000033  Westminster     ltla 2022-02-21                   58767   \n",
       "218888  E09000033  Westminster     ltla 2022-02-22                   58913   \n",
       "218889  E09000033  Westminster     ltla 2022-02-23                   59035   \n",
       "218890  E09000033  Westminster     ltla 2022-02-24                   59072   \n",
       "\n",
       "        cumCasesBySpecimenDateRate  newCasesBySpecimenDate  dateidx  \n",
       "0                              1.1                     1.0       75  \n",
       "1                              1.1                     0.0       76  \n",
       "2                              2.1                     1.0       77  \n",
       "3                              2.1                     0.0       78  \n",
       "4                              3.2                     1.0       79  \n",
       "...                            ...                     ...      ...  \n",
       "218886                     21725.6                   137.0      781  \n",
       "218887                     21777.8                   141.0      782  \n",
       "218888                     21831.9                   146.0      783  \n",
       "218889                     21877.1                   122.0      784  \n",
       "218890                     21890.8                    37.0      785  \n",
       "\n",
       "[218891 rows x 8 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd \n",
    "\n",
    "path = 'raw/covidinfo/ltla_2022-02-25.csv'\n",
    "df = pd.read_csv(path)\n",
    "df['date'] = pd.to_datetime(df['date'])\n",
    "df = df.sort_values(['areaCode', 'date']).reset_index(drop = True)\n",
    "ltla_covid_df = df[df['areaCode'].isin(ltla_df['areaCode'])].reset_index(drop = True)\n",
    "ltla_covid_df['dateidx'] = ltla_covid_df['date'].apply(lambda x: (x-pd.to_datetime('2020-01-01')).days)\n",
    "ltla_covid_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cf045d54",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.080257Z",
     "start_time": "2022-05-10T04:12:35.069916Z"
    }
   },
   "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>areaName</th>\n",
       "      <th>areaidx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Middlesbrough</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Redcar and Cleveland</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Stockton-on-Tees</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Darlington</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>300</th>\n",
       "      <td>Sutton</td>\n",
       "      <td>301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>301</th>\n",
       "      <td>Tower Hamlets</td>\n",
       "      <td>302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>302</th>\n",
       "      <td>Waltham Forest</td>\n",
       "      <td>303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303</th>\n",
       "      <td>Wandsworth</td>\n",
       "      <td>304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>304</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>305</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>305 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 areaName  areaidx\n",
       "0              Hartlepool        1\n",
       "1           Middlesbrough        2\n",
       "2    Redcar and Cleveland        3\n",
       "3        Stockton-on-Tees        4\n",
       "4              Darlington        5\n",
       "..                    ...      ...\n",
       "300                Sutton      301\n",
       "301         Tower Hamlets      302\n",
       "302        Waltham Forest      303\n",
       "303            Wandsworth      304\n",
       "304           Westminster      305\n",
       "\n",
       "[305 rows x 2 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "areaName_areaidx_df = ltla_df[['areaName', 'areaidx']]\n",
    "areaName_areaidx_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59197a19",
   "metadata": {},
   "source": [
    "# full data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "28028648",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.250843Z",
     "start_time": "2022-05-10T04:12:35.082027Z"
    }
   },
   "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>areaidx</th>\n",
       "      <th>areaCode</th>\n",
       "      <th>areaName</th>\n",
       "      <th>areaType</th>\n",
       "      <th>NHSER21NM</th>\n",
       "      <th>dateidx</th>\n",
       "      <th>date</th>\n",
       "      <th>newCasesRate</th>\n",
       "      <th>cumCasesBySpecimenDate</th>\n",
       "      <th>cumCasesBySpecimenDateRate</th>\n",
       "      <th>...</th>\n",
       "      <th>old_dependency_ratio</th>\n",
       "      <th>young_dependency_ratio</th>\n",
       "      <th>popu_all</th>\n",
       "      <th>Median age</th>\n",
       "      <th>People per sq. km</th>\n",
       "      <th>Average Score</th>\n",
       "      <th>Area (sq km)</th>\n",
       "      <th>male_avg_age</th>\n",
       "      <th>female_avg_age</th>\n",
       "      <th>female_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>75</td>\n",
       "      <td>2020-03-16</td>\n",
       "      <td>0.106569</td>\n",
       "      <td>1</td>\n",
       "      <td>1.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.288634</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>76</td>\n",
       "      <td>2020-03-17</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>1.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.288634</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>77</td>\n",
       "      <td>2020-03-18</td>\n",
       "      <td>0.106569</td>\n",
       "      <td>2</td>\n",
       "      <td>2.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.288634</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>78</td>\n",
       "      <td>2020-03-19</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>2.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.288634</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>79</td>\n",
       "      <td>2020-03-20</td>\n",
       "      <td>0.106569</td>\n",
       "      <td>3</td>\n",
       "      <td>3.2</td>\n",
       "      <td>...</td>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.288634</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218886</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>781</td>\n",
       "      <td>2022-02-20</td>\n",
       "      <td>5.076932</td>\n",
       "      <td>58626</td>\n",
       "      <td>21725.6</td>\n",
       "      <td>...</td>\n",
       "      <td>0.175793</td>\n",
       "      <td>0.223543</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218887</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>782</td>\n",
       "      <td>2022-02-21</td>\n",
       "      <td>5.225164</td>\n",
       "      <td>58767</td>\n",
       "      <td>21777.8</td>\n",
       "      <td>...</td>\n",
       "      <td>0.175793</td>\n",
       "      <td>0.223543</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218888</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>783</td>\n",
       "      <td>2022-02-22</td>\n",
       "      <td>5.410453</td>\n",
       "      <td>58913</td>\n",
       "      <td>21831.9</td>\n",
       "      <td>...</td>\n",
       "      <td>0.175793</td>\n",
       "      <td>0.223543</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218889</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>784</td>\n",
       "      <td>2022-02-23</td>\n",
       "      <td>4.521064</td>\n",
       "      <td>59035</td>\n",
       "      <td>21877.1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.175793</td>\n",
       "      <td>0.223543</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218890</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>785</td>\n",
       "      <td>2022-02-24</td>\n",
       "      <td>1.371142</td>\n",
       "      <td>59072</td>\n",
       "      <td>21890.8</td>\n",
       "      <td>...</td>\n",
       "      <td>0.175793</td>\n",
       "      <td>0.223543</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>218891 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        areaidx   areaCode     areaName areaType                 NHSER21NM  \\\n",
       "0             1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "1             1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "2             1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "3             1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "4             1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "...         ...        ...          ...      ...                       ...   \n",
       "218886      305  E09000033  Westminster     ltla                    London   \n",
       "218887      305  E09000033  Westminster     ltla                    London   \n",
       "218888      305  E09000033  Westminster     ltla                    London   \n",
       "218889      305  E09000033  Westminster     ltla                    London   \n",
       "218890      305  E09000033  Westminster     ltla                    London   \n",
       "\n",
       "        dateidx       date  newCasesRate  cumCasesBySpecimenDate  \\\n",
       "0            75 2020-03-16      0.106569                       1   \n",
       "1            76 2020-03-17      0.000000                       1   \n",
       "2            77 2020-03-18      0.106569                       2   \n",
       "3            78 2020-03-19      0.000000                       2   \n",
       "4            79 2020-03-20      0.106569                       3   \n",
       "...         ...        ...           ...                     ...   \n",
       "218886      781 2022-02-20      5.076932                   58626   \n",
       "218887      782 2022-02-21      5.225164                   58767   \n",
       "218888      783 2022-02-22      5.410453                   58913   \n",
       "218889      784 2022-02-23      4.521064                   59035   \n",
       "218890      785 2022-02-24      1.371142                   59072   \n",
       "\n",
       "        cumCasesBySpecimenDateRate  ...  old_dependency_ratio  \\\n",
       "0                              1.1  ...              0.313702   \n",
       "1                              1.1  ...              0.313702   \n",
       "2                              2.1  ...              0.313702   \n",
       "3                              2.1  ...              0.313702   \n",
       "4                              3.2  ...              0.313702   \n",
       "...                            ...  ...                   ...   \n",
       "218886                     21725.6  ...              0.175793   \n",
       "218887                     21777.8  ...              0.175793   \n",
       "218888                     21831.9  ...              0.175793   \n",
       "218889                     21877.1  ...              0.175793   \n",
       "218890                     21890.8  ...              0.175793   \n",
       "\n",
       "        young_dependency_ratio  popu_all  Median age  People per sq. km  \\\n",
       "0                     0.288634     93836        41.7                999   \n",
       "1                     0.288634     93836        41.7                999   \n",
       "2                     0.288634     93836        41.7                999   \n",
       "3                     0.288634     93836        41.7                999   \n",
       "4                     0.288634     93836        41.7                999   \n",
       "...                        ...       ...         ...                ...   \n",
       "218886                0.223543    269848        36.3              12167   \n",
       "218887                0.223543    269848        36.3              12167   \n",
       "218888                0.223543    269848        36.3              12167   \n",
       "218889                0.223543    269848        36.3              12167   \n",
       "218890                0.223543    269848        36.3              12167   \n",
       "\n",
       "        Average Score  Area (sq km)  male_avg_age  female_avg_age  female_rate  \n",
       "0                35.0       93.7125     40.157418       42.190274     0.509730  \n",
       "1                35.0       93.7125     40.157418       42.190274     0.509730  \n",
       "2                35.0       93.7125     40.157418       42.190274     0.509730  \n",
       "3                35.0       93.7125     40.157418       42.190274     0.509730  \n",
       "4                35.0       93.7125     40.157418       42.190274     0.509730  \n",
       "...               ...           ...           ...             ...          ...  \n",
       "218886           20.3       21.4826     36.918846       38.427939     0.472677  \n",
       "218887           20.3       21.4826     36.918846       38.427939     0.472677  \n",
       "218888           20.3       21.4826     36.918846       38.427939     0.472677  \n",
       "218889           20.3       21.4826     36.918846       38.427939     0.472677  \n",
       "218890           20.3       21.4826     36.918846       38.427939     0.472677  \n",
       "\n",
       "[218891 rows x 21 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_data_df = pd.merge(ltla_covid_df, ltla_df[['areaCode', 'NHSER21NM',\n",
    "       'old_dependency_ratio', 'young_dependency_ratio', 'popu_all',\n",
    "       'Median age', 'People per sq. km', 'Average Score', 'Area (sq km)',\n",
    "       'male_avg_age', 'female_avg_age', 'female_rate', 'areaidx']], on = 'areaCode')\n",
    "\n",
    "full_data_df['newCasesRate'] = full_data_df['newCasesBySpecimenDate'] / full_data_df['popu_all'] * 10000\n",
    "\n",
    "cols = [# areaidx\n",
    "        'areaidx', 'areaCode', 'areaName', 'areaType', 'NHSER21NM',\n",
    "        # date\n",
    "        'dateidx', 'date', \n",
    "        # y\n",
    "        'newCasesRate', 'cumCasesBySpecimenDate',\n",
    "        'cumCasesBySpecimenDateRate', 'newCasesBySpecimenDate', \n",
    "        # features\n",
    "        'old_dependency_ratio', 'young_dependency_ratio', 'popu_all',\n",
    "       'Median age', 'People per sq. km', 'Average Score', 'Area (sq km)',\n",
    "       'male_avg_age', 'female_avg_age', 'female_rate', \n",
    "       ]\n",
    "\n",
    "full_data_df = full_data_df[cols]\n",
    "full_data_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b596f99d",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.255678Z",
     "start_time": "2022-05-10T04:12:35.252999Z"
    }
   },
   "outputs": [],
   "source": [
    "def convert_dateidx_to_date(dateidx):\n",
    "    return (pd.to_timedelta(dateidx, unit = 'D') + pd.to_datetime('2020-01-01')).date()\n",
    "\n",
    "# dateidx = 10\n",
    "# convert_dateidx_to_date(dateidx)\n",
    "\n",
    "\n",
    "def convert_date_to_dateidx(date):\n",
    "    return (pd.to_datetime(date) - pd.to_datetime('2020-01-01')).days\n",
    "\n",
    "\n",
    "# date = '2021-5-1'\n",
    "# convert_date_to_dateidx(date)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "365fe8fa",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.260170Z",
     "start_time": "2022-05-10T04:12:35.257125Z"
    }
   },
   "outputs": [],
   "source": [
    "# areaName_areaidx_df#.query(f\"~({unit_col}=={treat_idx})\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "54a7359c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.262942Z",
     "start_time": "2022-05-10T04:12:35.261220Z"
    }
   },
   "outputs": [],
   "source": [
    "# areaName_areaidx_df.query(f'~(areaName==Liverpool)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7c6facb6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-10T04:12:35.279819Z",
     "start_time": "2022-05-10T04:12:35.264028Z"
    }
   },
   "outputs": [],
   "source": [
    "def convert_areaName_to_areaidx(areaName):\n",
    "    df = areaName_areaidx_df\n",
    "    return df[df['areaName'] == areaName].iloc[0]['areaidx']\n",
    "    \n",
    "# areaName = 'Liverpool'\n",
    "# areaidx = convert_areaName_to_areaidx(areaName)\n",
    "# print(areaidx)\n",
    "\n",
    "def convert_areaidx_to_areaName(areaidx):\n",
    "    df = areaName_areaidx_df\n",
    "    return df[df['areaidx'] == areaidx].iloc[0]['areaName']\n",
    "\n",
    "# areaidx = 249\n",
    "# areaName = convert_areaidx_to_areaName(areaidx)\n",
    "# print(areaName)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d190cbc6",
   "metadata": {},
   "source": [
    "# Game Info"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "91220815",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:05.973696Z",
     "start_time": "2022-04-14T22:59:05.951478Z"
    }
   },
   "outputs": [],
   "source": [
    "ltla_game_df = pd.read_csv('data/ltla_game_df.csv')\n",
    "ltla_game_df['date'] = pd.to_datetime(ltla_game_df['date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "8162acf3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:17.201444Z",
     "start_time": "2022-04-14T22:59:16.463096Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 3600x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_covid_soccer(city_covid_df, city_game_df, y_col):\n",
    "    fig, ax = plt.subplots(figsize=(50, 10))\n",
    "    df = city_covid_df# [start:end]\n",
    "    df[[# 'cumCasesBySpecimenDate',\n",
    "        # 'cumCasesBySpecimenDateRate', \n",
    "        y_col,\n",
    "       ]].plot(ax = ax)\n",
    "    xposition = city_game_df['date']\n",
    "    for xc in xposition:\n",
    "        ax.axvline(x=xc, color='grey', linestyle='--')\n",
    "    xposition = city_game_df[city_game_df['win'] == 1]['date']\n",
    "    for xc in xposition:\n",
    "        ax.axvline(x = xc, color='green', linestyle='--')\n",
    "    xposition = city_game_df[city_game_df['host'] == 1]\n",
    "    xposition = xposition[xposition['win'] == 1]['date']\n",
    "    for xc in xposition:\n",
    "        ax.axvline(x = xc, color='red', linestyle='--')\n",
    "        \n",
    "        \n",
    "\n",
    "city_name = 'Liverpool'\n",
    "y_col = 'newCasesRate' # 10k\n",
    "city_covid_df = full_data_df[full_data_df['areaName'] == city_name].reset_index(drop = True)\n",
    "city_covid_df = city_covid_df.set_index('date')\n",
    "city_game_df = ltla_game_df[ltla_game_df['city'] == city_name].reset_index(drop = True)\n",
    "plot_covid_soccer(city_covid_df, city_game_df, y_col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "99c66c2f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:19.222642Z",
     "start_time": "2022-04-14T22:59:19.135903Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-09-28 ---- 2019-08-24 2021-11-20\n",
      "2021-11-20 ---- 2020-09-28 2019-09-14\n",
      "2021-04-24 ---- 2019-09-14 2021-12-16\n",
      "2021-12-16 ---- 2021-04-24 2019-10-05\n",
      "2020-11-22 ---- 2019-10-05 2019-10-27\n",
      "2020-12-16 ---- 2019-10-27 2019-11-10\n",
      "2021-02-07 ---- 2019-11-10 2021-10-03\n",
      "2021-10-03 ---- 2021-02-07 2019-11-30\n",
      "2021-02-03 ---- 2019-11-30 2021-10-30\n",
      "2021-10-30 ---- 2021-02-03 2019-12-04\n",
      "2021-02-20 ---- 2019-12-04 2021-02-20\n",
      "2019-12-29 ---- 2019-12-14 2020-12-06\n",
      "2020-12-06 ---- 2019-12-29 2020-01-02\n",
      "2020-10-24 ---- 2020-01-02 2020-01-19\n",
      "2021-01-17 ---- 2020-01-19 2020-02-01\n",
      "2021-05-08 ---- 2020-02-01 2021-11-27\n",
      "2021-11-27 ---- 2021-05-08 2020-02-24\n",
      "2020-10-31 ---- 2020-02-24 2020-03-07\n",
      "2020-06-24 ---- 2020-03-07 2021-05-23\n",
      "2021-05-23 ---- 2020-06-24 2021-09-18\n",
      "2021-09-18 ---- 2021-05-23 2020-07-05\n",
      "2021-04-10 ---- 2020-07-05 2021-12-11\n",
      "2021-12-11 ---- 2021-04-10 2020-07-11\n",
      "2021-01-21 ---- 2020-07-11 2021-08-21\n",
      "2021-08-21 ---- 2021-01-21 2020-07-22\n",
      "2021-03-04 ---- 2020-07-22 2021-08-28\n",
      "2021-08-28 ---- 2021-03-04 2020-09-12\n",
      "2020-12-27 ---- 2020-09-12 2021-03-07\n",
      "2021-03-07 ---- 2020-12-27 2022-01-16\n",
      "2022-01-16 ---- 2021-03-07 2020-01-18\n",
      "2021-05-09 ---- 2020-01-18 2020-01-29\n",
      "2021-01-31 ---- 2020-01-29 2021-11-07\n",
      "2021-11-07 ---- 2021-01-31 2020-02-01\n",
      "2020-02-29 ---- 2020-02-01 2021-10-16\n",
      "2021-10-16 ---- 2020-02-29 2019-08-10\n",
      "2020-09-26 ---- 2019-08-10 2021-12-12\n",
      "2021-12-12 ---- 2020-09-26 2019-11-23\n",
      "2020-12-19 ---- 2019-11-23 2022-01-23\n",
      "2022-01-23 ---- 2020-12-19 2019-10-05\n",
      "2020-12-05 ---- 2019-10-05 2019-08-31\n",
      "2021-05-19 ---- 2019-08-31 2020-07-06\n",
      "2020-09-13 ---- 2020-07-06 2020-01-11\n",
      "2021-01-28 ---- 2020-01-11 2021-12-19\n",
      "2021-12-19 ---- 2021-01-28 2019-12-28\n",
      "2020-11-01 ---- 2019-12-28 2020-07-26\n",
      "2020-12-30 ---- 2020-07-26 2020-12-16\n",
      "2021-02-06 ---- 2019-12-15 2021-10-02\n",
      "2021-10-02 ---- 2021-02-06 2019-10-20\n",
      "2021-05-13 ---- 2019-10-20 2021-10-24\n",
      "2021-10-24 ---- 2021-05-13 2020-02-23\n",
      "2021-04-23 ---- 2020-02-23 2020-07-15\n",
      "2021-04-03 ---- 2020-07-15 2019-11-09\n",
      "2020-10-25 ---- 2019-11-09 2019-08-17\n",
      "2021-01-04 ---- 2019-08-17 2019-08-23\n",
      "2021-05-13 ---- 2019-08-23 2021-09-18\n",
      "2021-09-18 ---- 2021-05-13 2019-11-02\n",
      "2020-10-04 ---- 2019-11-02 2019-11-23\n",
      "2021-09-25 ---- 2019-11-23 2019-12-21\n",
      "2020-12-19 ---- 2019-12-21 2021-12-06\n",
      "2021-12-06 ---- 2020-12-19 2020-01-21\n",
      "2021-01-30 ---- 2020-01-21 2020-07-01\n",
      "2021-04-16 ---- 2019-11-03 2021-11-07\n",
      "2021-11-07 ---- 2021-04-16 2019-09-28\n",
      "2021-02-17 ---- 2019-09-28 2020-01-11\n",
      "2020-10-03 ---- 2020-01-11 2022-01-02\n",
      "2022-01-02 ---- 2020-10-03 2019-08-17\n",
      "2021-10-23 ---- 2019-08-17 2019-09-01\n",
      "2021-05-19 ---- 2019-09-01 2020-03-01\n",
      "2020-11-07 ---- 2020-03-01 2020-07-09\n",
      "2021-03-01 ---- 2020-07-09 2021-08-14\n",
      "2021-08-14 ---- 2021-03-01 2019-10-19\n",
      "2021-01-01 ---- 2019-10-19 2021-10-17\n",
      "2021-10-17 ---- 2021-01-01 2020-07-26\n",
      "2021-04-05 ---- 2020-02-08 2020-07-16\n",
      "2021-05-01 ---- 2020-07-16 2022-01-22\n",
      "2022-01-22 ---- 2021-05-01 2019-12-26\n",
      "2021-03-13 ---- 2019-12-26 2021-09-13\n",
      "2021-09-13 ---- 2021-03-13 2019-12-07\n",
      "2020-12-12 ---- 2019-12-07 2020-11-28\n",
      "2021-02-14 ---- 2020-09-19 2020-06-21\n",
      "2020-10-17 ---- 2020-06-21 2020-10-17\n",
      "2021-12-01 ---- 2020-10-17 2021-12-01\n",
      "2021-04-12 ---- 2019-10-26 2021-08-28\n",
      "2021-08-28 ---- 2021-04-12 2020-07-08\n",
      "2020-11-28 ---- 2020-07-08 2020-01-01\n",
      "2021-05-23 ---- 2020-01-01 2021-11-21\n",
      "2021-11-21 ---- 2021-05-23 2020-07-02\n",
      "2020-11-08 ---- 2020-07-02 2020-07-12\n",
      "2021-01-12 ---- 2020-07-12 2021-11-01\n",
      "2021-11-01 ---- 2021-01-12 2020-01-23\n",
      "2021-03-15 ---- 2020-01-23 2021-12-04\n",
      "2021-12-04 ---- 2021-03-15 2019-09-15\n",
      "2019-12-07 ---- 2019-09-15 2020-03-08\n",
      "2020-03-08 ---- 2019-12-07 2021-03-08\n",
      "2021-03-08 ---- 2020-03-08 2021-12-16\n",
      "2021-12-16 ---- 2021-03-08 2019-09-22\n",
      "2020-09-20 ---- 2019-09-22 2022-01-02\n",
      "2022-01-02 ---- 2020-09-20 2021-11-28\n",
      "2022-01-15 ---- 2020-06-24 2020-02-15\n",
      "2021-08-14 ---- 2020-02-15 2020-07-20\n",
      "2020-12-26 ---- 2020-07-20 2019-09-28\n",
      "2021-02-28 ---- 2019-09-28 2020-11-22\n",
      "2020-12-13 ---- 2020-11-22 2021-02-03\n",
      "2021-02-03 ---- 2020-12-13 2021-08-21\n",
      "2021-08-21 ---- 2021-02-03 2021-04-19\n",
      "2021-09-12 ---- 2021-04-19 2021-03-04\n"
     ]
    }
   ],
   "source": [
    "required_gap = 7 # days\n",
    "\n",
    "gaps = pd.to_timedelta(required_gap, unit = 'D')\n",
    "\n",
    "for idx, row in city_game_df.iterrows():\n",
    "    if idx <= 1 or idx == len(city_game_df) - 1: continue\n",
    "    date = row['date']\n",
    "    last_game = city_game_df.loc[idx - 1]['date']\n",
    "    next_game = city_game_df.loc[idx + 1]['date']\n",
    "    # if date - gaps > last_game and date + gaps < next_game:\n",
    "    if date - gaps > last_game:\n",
    "    # if date + gaps < next_game:\n",
    "        print(date.date(), '----', last_game.date(), next_game.date())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a15de0b",
   "metadata": {},
   "source": [
    "# City - Date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9e750d73",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:24.976160Z",
     "start_time": "2022-04-14T22:59:24.967314Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "areaidx: 249, dateidx: 255\n",
      "gap is 7\n"
     ]
    }
   ],
   "source": [
    "treat_areaName =\"Liverpool\" #  'Leeds' # \"Liverpool\"\n",
    "treat_date = '2020-09-12' # '2020-06-26' # '2020-11-07' #'2020-06-26'\n",
    "gap = 7\n",
    "\n",
    "# argument: \n",
    "features = [] \n",
    "\n",
    "########################\n",
    "\n",
    "\n",
    "treat_areaidx = convert_areaName_to_areaidx(treat_areaName)\n",
    "treat_dateidx = convert_date_to_dateidx(treat_date)\n",
    "print('areaidx: {}, dateidx: {}'.format(treat_areaidx, treat_dateidx))\n",
    "print('gap is {}'.format(gap))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "604914c0",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:31.363373Z",
     "start_time": "2022-04-14T22:59:31.242496Z"
    }
   },
   "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>areaidx</th>\n",
       "      <th>areaCode</th>\n",
       "      <th>areaName</th>\n",
       "      <th>areaType</th>\n",
       "      <th>NHSER21NM</th>\n",
       "      <th>dateidx</th>\n",
       "      <th>date</th>\n",
       "      <th>newCasesRate</th>\n",
       "      <th>cumCasesBySpecimenDate</th>\n",
       "      <th>cumCasesBySpecimenDateRate</th>\n",
       "      <th>...</th>\n",
       "      <th>popu_all</th>\n",
       "      <th>Median age</th>\n",
       "      <th>People per sq. km</th>\n",
       "      <th>Average Score</th>\n",
       "      <th>Area (sq km)</th>\n",
       "      <th>male_avg_age</th>\n",
       "      <th>female_avg_age</th>\n",
       "      <th>female_rate</th>\n",
       "      <th>post_treatment</th>\n",
       "      <th>treat_unit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>248</td>\n",
       "      <td>2020-09-05</td>\n",
       "      <td>0.319707</td>\n",
       "      <td>705</td>\n",
       "      <td>751.3</td>\n",
       "      <td>...</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>249</td>\n",
       "      <td>2020-09-06</td>\n",
       "      <td>0.426276</td>\n",
       "      <td>709</td>\n",
       "      <td>755.6</td>\n",
       "      <td>...</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>250</td>\n",
       "      <td>2020-09-07</td>\n",
       "      <td>1.172258</td>\n",
       "      <td>720</td>\n",
       "      <td>767.3</td>\n",
       "      <td>...</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>251</td>\n",
       "      <td>2020-09-08</td>\n",
       "      <td>1.065689</td>\n",
       "      <td>730</td>\n",
       "      <td>778.0</td>\n",
       "      <td>...</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>E06000001</td>\n",
       "      <td>Hartlepool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North East and Yorkshire</td>\n",
       "      <td>252</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>0.745982</td>\n",
       "      <td>737</td>\n",
       "      <td>785.4</td>\n",
       "      <td>...</td>\n",
       "      <td>93836</td>\n",
       "      <td>41.7</td>\n",
       "      <td>999</td>\n",
       "      <td>35.0</td>\n",
       "      <td>93.7125</td>\n",
       "      <td>40.157418</td>\n",
       "      <td>42.190274</td>\n",
       "      <td>0.509730</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4165</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>258</td>\n",
       "      <td>2020-09-15</td>\n",
       "      <td>0.444695</td>\n",
       "      <td>1073</td>\n",
       "      <td>397.6</td>\n",
       "      <td>...</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4166</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>259</td>\n",
       "      <td>2020-09-16</td>\n",
       "      <td>0.370579</td>\n",
       "      <td>1083</td>\n",
       "      <td>401.3</td>\n",
       "      <td>...</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4167</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>260</td>\n",
       "      <td>2020-09-17</td>\n",
       "      <td>0.259405</td>\n",
       "      <td>1090</td>\n",
       "      <td>403.9</td>\n",
       "      <td>...</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4168</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>261</td>\n",
       "      <td>2020-09-18</td>\n",
       "      <td>0.518811</td>\n",
       "      <td>1104</td>\n",
       "      <td>409.1</td>\n",
       "      <td>...</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4169</th>\n",
       "      <td>305</td>\n",
       "      <td>E09000033</td>\n",
       "      <td>Westminster</td>\n",
       "      <td>ltla</td>\n",
       "      <td>London</td>\n",
       "      <td>262</td>\n",
       "      <td>2020-09-19</td>\n",
       "      <td>0.111174</td>\n",
       "      <td>1107</td>\n",
       "      <td>410.2</td>\n",
       "      <td>...</td>\n",
       "      <td>269848</td>\n",
       "      <td>36.3</td>\n",
       "      <td>12167</td>\n",
       "      <td>20.3</td>\n",
       "      <td>21.4826</td>\n",
       "      <td>36.918846</td>\n",
       "      <td>38.427939</td>\n",
       "      <td>0.472677</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4170 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      areaidx   areaCode     areaName areaType                 NHSER21NM  \\\n",
       "0           1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "1           1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "2           1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "3           1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "4           1  E06000001   Hartlepool     ltla  North East and Yorkshire   \n",
       "...       ...        ...          ...      ...                       ...   \n",
       "4165      305  E09000033  Westminster     ltla                    London   \n",
       "4166      305  E09000033  Westminster     ltla                    London   \n",
       "4167      305  E09000033  Westminster     ltla                    London   \n",
       "4168      305  E09000033  Westminster     ltla                    London   \n",
       "4169      305  E09000033  Westminster     ltla                    London   \n",
       "\n",
       "      dateidx       date  newCasesRate  cumCasesBySpecimenDate  \\\n",
       "0         248 2020-09-05      0.319707                     705   \n",
       "1         249 2020-09-06      0.426276                     709   \n",
       "2         250 2020-09-07      1.172258                     720   \n",
       "3         251 2020-09-08      1.065689                     730   \n",
       "4         252 2020-09-09      0.745982                     737   \n",
       "...       ...        ...           ...                     ...   \n",
       "4165      258 2020-09-15      0.444695                    1073   \n",
       "4166      259 2020-09-16      0.370579                    1083   \n",
       "4167      260 2020-09-17      0.259405                    1090   \n",
       "4168      261 2020-09-18      0.518811                    1104   \n",
       "4169      262 2020-09-19      0.111174                    1107   \n",
       "\n",
       "      cumCasesBySpecimenDateRate  ...  popu_all  Median age  \\\n",
       "0                          751.3  ...     93836        41.7   \n",
       "1                          755.6  ...     93836        41.7   \n",
       "2                          767.3  ...     93836        41.7   \n",
       "3                          778.0  ...     93836        41.7   \n",
       "4                          785.4  ...     93836        41.7   \n",
       "...                          ...  ...       ...         ...   \n",
       "4165                       397.6  ...    269848        36.3   \n",
       "4166                       401.3  ...    269848        36.3   \n",
       "4167                       403.9  ...    269848        36.3   \n",
       "4168                       409.1  ...    269848        36.3   \n",
       "4169                       410.2  ...    269848        36.3   \n",
       "\n",
       "      People per sq. km  Average Score  Area (sq km)  male_avg_age  \\\n",
       "0                   999           35.0       93.7125     40.157418   \n",
       "1                   999           35.0       93.7125     40.157418   \n",
       "2                   999           35.0       93.7125     40.157418   \n",
       "3                   999           35.0       93.7125     40.157418   \n",
       "4                   999           35.0       93.7125     40.157418   \n",
       "...                 ...            ...           ...           ...   \n",
       "4165              12167           20.3       21.4826     36.918846   \n",
       "4166              12167           20.3       21.4826     36.918846   \n",
       "4167              12167           20.3       21.4826     36.918846   \n",
       "4168              12167           20.3       21.4826     36.918846   \n",
       "4169              12167           20.3       21.4826     36.918846   \n",
       "\n",
       "      female_avg_age  female_rate  post_treatment  treat_unit  \n",
       "0          42.190274     0.509730           False       False  \n",
       "1          42.190274     0.509730           False       False  \n",
       "2          42.190274     0.509730           False       False  \n",
       "3          42.190274     0.509730           False       False  \n",
       "4          42.190274     0.509730           False       False  \n",
       "...              ...          ...             ...         ...  \n",
       "4165       38.427939     0.472677            True       False  \n",
       "4166       38.427939     0.472677            True       False  \n",
       "4167       38.427939     0.472677            True       False  \n",
       "4168       38.427939     0.472677            True       False  \n",
       "4169       38.427939     0.472677            True       False  \n",
       "\n",
       "[4170 rows x 23 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# filter out the unwanted cities\n",
    "game_ltla_list = [i for i in ltla_game_df['city'].unique() if i != treat_areaName]\n",
    "neighours = ltla_neighbours_df[ltla_neighbours_df['LAD21NM'] == treat_areaName]['neighboors'].values[0]\n",
    "\n",
    "data = full_data_df[- full_data_df['areaName'].isin(game_ltla_list + neighours)].reset_index(drop = True)\n",
    "# filter out the unwanted dates\n",
    "data = data[data['dateidx'].apply(lambda x: abs(x - treat_dateidx) <= gap)].reset_index(drop = True)\n",
    "\n",
    "\n",
    "# Donor Pool. \n",
    "data['post_treatment'] = (data['dateidx'] >= treat_dateidx)\n",
    "data['treat_unit'] = (data['areaidx'] == treat_areaidx)\n",
    "data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ba9e6a26",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:34.723016Z",
     "start_time": "2022-04-14T22:59:34.681953Z"
    },
    "scrolled": true
   },
   "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>areaidx</th>\n",
       "      <th>areaCode</th>\n",
       "      <th>areaName</th>\n",
       "      <th>areaType</th>\n",
       "      <th>NHSER21NM</th>\n",
       "      <th>dateidx</th>\n",
       "      <th>date</th>\n",
       "      <th>newCasesRate</th>\n",
       "      <th>cumCasesBySpecimenDate</th>\n",
       "      <th>cumCasesBySpecimenDateRate</th>\n",
       "      <th>...</th>\n",
       "      <th>popu_all</th>\n",
       "      <th>Median age</th>\n",
       "      <th>People per sq. km</th>\n",
       "      <th>Average Score</th>\n",
       "      <th>Area (sq km)</th>\n",
       "      <th>male_avg_age</th>\n",
       "      <th>female_avg_age</th>\n",
       "      <th>female_rate</th>\n",
       "      <th>post_treatment</th>\n",
       "      <th>treat_unit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3510</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>248</td>\n",
       "      <td>2020-09-05</td>\n",
       "      <td>1.258807</td>\n",
       "      <td>3160</td>\n",
       "      <td>631.4</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3511</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>249</td>\n",
       "      <td>2020-09-06</td>\n",
       "      <td>0.699337</td>\n",
       "      <td>3195</td>\n",
       "      <td>638.4</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3512</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>250</td>\n",
       "      <td>2020-09-07</td>\n",
       "      <td>1.338731</td>\n",
       "      <td>3262</td>\n",
       "      <td>651.8</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3513</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>251</td>\n",
       "      <td>2020-09-08</td>\n",
       "      <td>1.178882</td>\n",
       "      <td>3321</td>\n",
       "      <td>663.6</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3514</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>252</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>1.858238</td>\n",
       "      <td>3414</td>\n",
       "      <td>682.2</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3515</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>253</td>\n",
       "      <td>2020-09-10</td>\n",
       "      <td>1.838257</td>\n",
       "      <td>3506</td>\n",
       "      <td>700.5</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3516</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>254</td>\n",
       "      <td>2020-09-11</td>\n",
       "      <td>1.738352</td>\n",
       "      <td>3593</td>\n",
       "      <td>717.9</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3517</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>255</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>1.558523</td>\n",
       "      <td>3671</td>\n",
       "      <td>733.5</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3518</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>256</td>\n",
       "      <td>2020-09-13</td>\n",
       "      <td>1.598485</td>\n",
       "      <td>3751</td>\n",
       "      <td>749.5</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3519</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>257</td>\n",
       "      <td>2020-09-14</td>\n",
       "      <td>1.658428</td>\n",
       "      <td>3834</td>\n",
       "      <td>766.1</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3520</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>258</td>\n",
       "      <td>2020-09-15</td>\n",
       "      <td>1.838257</td>\n",
       "      <td>3926</td>\n",
       "      <td>784.5</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3521</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>259</td>\n",
       "      <td>2020-09-16</td>\n",
       "      <td>2.437689</td>\n",
       "      <td>4048</td>\n",
       "      <td>808.8</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3522</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>260</td>\n",
       "      <td>2020-09-17</td>\n",
       "      <td>3.216950</td>\n",
       "      <td>4209</td>\n",
       "      <td>841.0</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3523</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>261</td>\n",
       "      <td>2020-09-18</td>\n",
       "      <td>3.356818</td>\n",
       "      <td>4377</td>\n",
       "      <td>874.6</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3524</th>\n",
       "      <td>249</td>\n",
       "      <td>E08000012</td>\n",
       "      <td>Liverpool</td>\n",
       "      <td>ltla</td>\n",
       "      <td>North West</td>\n",
       "      <td>262</td>\n",
       "      <td>2020-09-19</td>\n",
       "      <td>3.316856</td>\n",
       "      <td>4543</td>\n",
       "      <td>907.7</td>\n",
       "      <td>...</td>\n",
       "      <td>500474</td>\n",
       "      <td>34.8</td>\n",
       "      <td>4453</td>\n",
       "      <td>42.4</td>\n",
       "      <td>111.8357</td>\n",
       "      <td>36.823739</td>\n",
       "      <td>38.438335</td>\n",
       "      <td>0.499682</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      areaidx   areaCode   areaName areaType   NHSER21NM  dateidx       date  \\\n",
       "3510      249  E08000012  Liverpool     ltla  North West      248 2020-09-05   \n",
       "3511      249  E08000012  Liverpool     ltla  North West      249 2020-09-06   \n",
       "3512      249  E08000012  Liverpool     ltla  North West      250 2020-09-07   \n",
       "3513      249  E08000012  Liverpool     ltla  North West      251 2020-09-08   \n",
       "3514      249  E08000012  Liverpool     ltla  North West      252 2020-09-09   \n",
       "3515      249  E08000012  Liverpool     ltla  North West      253 2020-09-10   \n",
       "3516      249  E08000012  Liverpool     ltla  North West      254 2020-09-11   \n",
       "3517      249  E08000012  Liverpool     ltla  North West      255 2020-09-12   \n",
       "3518      249  E08000012  Liverpool     ltla  North West      256 2020-09-13   \n",
       "3519      249  E08000012  Liverpool     ltla  North West      257 2020-09-14   \n",
       "3520      249  E08000012  Liverpool     ltla  North West      258 2020-09-15   \n",
       "3521      249  E08000012  Liverpool     ltla  North West      259 2020-09-16   \n",
       "3522      249  E08000012  Liverpool     ltla  North West      260 2020-09-17   \n",
       "3523      249  E08000012  Liverpool     ltla  North West      261 2020-09-18   \n",
       "3524      249  E08000012  Liverpool     ltla  North West      262 2020-09-19   \n",
       "\n",
       "      newCasesRate  cumCasesBySpecimenDate  cumCasesBySpecimenDateRate  ...  \\\n",
       "3510      1.258807                    3160                       631.4  ...   \n",
       "3511      0.699337                    3195                       638.4  ...   \n",
       "3512      1.338731                    3262                       651.8  ...   \n",
       "3513      1.178882                    3321                       663.6  ...   \n",
       "3514      1.858238                    3414                       682.2  ...   \n",
       "3515      1.838257                    3506                       700.5  ...   \n",
       "3516      1.738352                    3593                       717.9  ...   \n",
       "3517      1.558523                    3671                       733.5  ...   \n",
       "3518      1.598485                    3751                       749.5  ...   \n",
       "3519      1.658428                    3834                       766.1  ...   \n",
       "3520      1.838257                    3926                       784.5  ...   \n",
       "3521      2.437689                    4048                       808.8  ...   \n",
       "3522      3.216950                    4209                       841.0  ...   \n",
       "3523      3.356818                    4377                       874.6  ...   \n",
       "3524      3.316856                    4543                       907.7  ...   \n",
       "\n",
       "      popu_all  Median age  People per sq. km  Average Score  Area (sq km)  \\\n",
       "3510    500474        34.8               4453           42.4      111.8357   \n",
       "3511    500474        34.8               4453           42.4      111.8357   \n",
       "3512    500474        34.8               4453           42.4      111.8357   \n",
       "3513    500474        34.8               4453           42.4      111.8357   \n",
       "3514    500474        34.8               4453           42.4      111.8357   \n",
       "3515    500474        34.8               4453           42.4      111.8357   \n",
       "3516    500474        34.8               4453           42.4      111.8357   \n",
       "3517    500474        34.8               4453           42.4      111.8357   \n",
       "3518    500474        34.8               4453           42.4      111.8357   \n",
       "3519    500474        34.8               4453           42.4      111.8357   \n",
       "3520    500474        34.8               4453           42.4      111.8357   \n",
       "3521    500474        34.8               4453           42.4      111.8357   \n",
       "3522    500474        34.8               4453           42.4      111.8357   \n",
       "3523    500474        34.8               4453           42.4      111.8357   \n",
       "3524    500474        34.8               4453           42.4      111.8357   \n",
       "\n",
       "      male_avg_age  female_avg_age  female_rate  post_treatment  treat_unit  \n",
       "3510     36.823739       38.438335     0.499682           False        True  \n",
       "3511     36.823739       38.438335     0.499682           False        True  \n",
       "3512     36.823739       38.438335     0.499682           False        True  \n",
       "3513     36.823739       38.438335     0.499682           False        True  \n",
       "3514     36.823739       38.438335     0.499682           False        True  \n",
       "3515     36.823739       38.438335     0.499682           False        True  \n",
       "3516     36.823739       38.438335     0.499682           False        True  \n",
       "3517     36.823739       38.438335     0.499682            True        True  \n",
       "3518     36.823739       38.438335     0.499682            True        True  \n",
       "3519     36.823739       38.438335     0.499682            True        True  \n",
       "3520     36.823739       38.438335     0.499682            True        True  \n",
       "3521     36.823739       38.438335     0.499682            True        True  \n",
       "3522     36.823739       38.438335     0.499682            True        True  \n",
       "3523     36.823739       38.438335     0.499682            True        True  \n",
       "3524     36.823739       38.438335     0.499682            True        True  \n",
       "\n",
       "[15 rows x 23 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data['areaidx'] == treat_areaidx]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b02a824b",
   "metadata": {},
   "source": [
    "# Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "585e7168",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:55.356921Z",
     "start_time": "2022-04-14T22:59:55.315519Z"
    },
    "scrolled": false
   },
   "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>groups</th>\n",
       "      <th>Liverpool</th>\n",
       "      <th>Other Cities</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-09-05</th>\n",
       "      <td>1.258807</td>\n",
       "      <td>0.282078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-06</th>\n",
       "      <td>0.699337</td>\n",
       "      <td>0.280492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-07</th>\n",
       "      <td>1.338731</td>\n",
       "      <td>0.493527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-08</th>\n",
       "      <td>1.178882</td>\n",
       "      <td>0.448162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-09</th>\n",
       "      <td>1.858238</td>\n",
       "      <td>0.396686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-10</th>\n",
       "      <td>1.838257</td>\n",
       "      <td>0.419399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-11</th>\n",
       "      <td>1.738352</td>\n",
       "      <td>0.375430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-12</th>\n",
       "      <td>1.558523</td>\n",
       "      <td>0.282299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-13</th>\n",
       "      <td>1.598485</td>\n",
       "      <td>0.234695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-14</th>\n",
       "      <td>1.658428</td>\n",
       "      <td>0.401182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-15</th>\n",
       "      <td>1.838257</td>\n",
       "      <td>0.405779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-16</th>\n",
       "      <td>2.437689</td>\n",
       "      <td>0.473890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-17</th>\n",
       "      <td>3.216950</td>\n",
       "      <td>0.484895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-18</th>\n",
       "      <td>3.356818</td>\n",
       "      <td>0.518740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-09-19</th>\n",
       "      <td>3.316856</td>\n",
       "      <td>0.516851</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "groups      Liverpool  Other Cities\n",
       "date                               \n",
       "2020-09-05   1.258807      0.282078\n",
       "2020-09-06   0.699337      0.280492\n",
       "2020-09-07   1.338731      0.493527\n",
       "2020-09-08   1.178882      0.448162\n",
       "2020-09-09   1.858238      0.396686\n",
       "2020-09-10   1.838257      0.419399\n",
       "2020-09-11   1.738352      0.375430\n",
       "2020-09-12   1.558523      0.282299\n",
       "2020-09-13   1.598485      0.234695\n",
       "2020-09-14   1.658428      0.401182\n",
       "2020-09-15   1.838257      0.405779\n",
       "2020-09-16   2.437689      0.473890\n",
       "2020-09-17   3.216950      0.484895\n",
       "2020-09-18   3.356818      0.518740\n",
       "2020-09-19   3.316856      0.516851"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "\n",
    "time_col = 'date'\n",
    "treat_col = 'treat_unit'\n",
    "y_col = 'newCasesRate'\n",
    "\n",
    "df = (data\n",
    " .assign(groups = np.where(data[treat_col], treat_areaName, \"Other Cities\"))\n",
    " .groupby([time_col, \"groups\"])[y_col]\n",
    " .mean()\n",
    " .reset_index()\n",
    " .pivot(time_col, 'groups', y_col)\n",
    ")\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "73325c33",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:56.284957Z",
     "start_time": "2022-04-14T22:59:56.282003Z"
    }
   },
   "outputs": [],
   "source": [
    "# date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f0869a6c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:56.939553Z",
     "start_time": "2022-04-14T22:59:56.775692Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1661d4c10>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAFkCAYAAABlzpwiAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAABk3klEQVR4nO3dd3wVVf7/8dcnhYSWUINIC71JDyAINlTsvbsqCiLYXVfXsqu7urp+d92f66prWVDEir27K2DFAoRepXdM6Akl/fz+mEkIIQkJ5GZukvfz8biPO3fuzNz3vQncT86cc8acc4iIiIhIeIgIOoCIiIiI7KfiTERERCSMqDgTERERCSMqzkRERETCiIozERERkTCi4kxEREQkjKg4E6nmzOx5M/tj0DlCwcw6m9lcM0s3s9sq+Ni7zaxdRR5TSmZmI8xsWtA5RMKBijORAJjZ5WY23cz2mFmqv3yTmVlFv5Zzboxz7pGKPm6omJkzsw5l3Pwe4GvnXH3n3L+O4DW/MbNRhdc55+o551Yd7jFLea01ZrbPL/5+NbMJZlavjPuGdQFjZseY2f/MbKuZHTSJppk1MrMP/N/7tWZ2ZRA5RcKdijORSmZmdwFPAX8HjgKaAWOA44BaAUaritoAi4IOcRjOcc7VA3oDfYD7go1TfmYWVczqbOBtYGQJuz0LZOH9zl8FPGdm3UOTUKTqUnEmUonMLB54GLjJOfeucy7deeY4565yzmX6251lZnPMLM3M1pvZnwodI9FvXRptZpvMbLOZ/a6U15xgZn/xl080sw1mdpffYrfZzK4rZd9vzOwRM/vBP3X4pZk1KfT8sWb2o5ntNLN5Znaiv/4kM1tQaLvJZjaz0OPvzez8MnxefzKzt81sov/6i8wsyX/uK+Ak4Bm/FaqTmcWY2RNmts7MUvxTurULHe88/zRompmtNLPTzexRYGih4zzjb1vQgmdm8X6GLX6Lzx/MLMJ/boSZTfNfd4eZrTazMw713gCcc78C/8Mr0vIz3utnSzezxWZ2gb++K/A8MMjPudNfX+p7LvJ5jvB/ls+Y2S4zW2pmwwo9H29m4/3fi41m9hcziyyy75Nmtg34UzHv5xfn3HiKKZjNrC5wEfBH59xu59w04GPg6hKy/t3/XOPL8FGKVCsqzkQq1yAgBvjoENvtAa4BGgBnAWOLKWZOAjoCpwG/N7NTypjhKCAeaIHXwvGsmTUsZfsrgeuABLyWvd8BmFkL4DPgL0Ajf/17ZtYU+BnoaGZNzCwa6AkcbWb1/cIhCfi+jHnPBd7C+yw+Bp4BcM6d7B/jFv8U5DLgcaATXrHTwX+PD/p5BwATgbv9Yx0PrHHOPVDkOLcUk+Fp/zNrB5yA97MpXNQOBH4BmgB/A8abHfoUtZm1BM4AVhRavRKvWIwH/gy8ZmbNnXNL8FpYf/JzNvC3L/E9l2Cg/xpNgIeA982skf/cBCDHP04fvN+tUUX2XYXX8vXood5fEZ2AHP/nlG8ecEDLmZlFmNl/8H5nTnPO7Srn64hUeSrORCpXE2Crcy4nf0Whlqd9ZnY8gHPuG+fcAudcnnNuPvAmXlFQ2J+dc3uccwuAl4ErypghG3jYOZftnPsc2A10LmX7l51zy5xz+/BOWfX21/8G+Nw597mfczKQDJzpbzsTrwDqh/cl/APeqdtjgeXOuW1lzDvNf41c4FWgV3Eb+cXQaOBO59x251w68Bhwub/JSOAl59xkP+9G59zSQ72433J0OXCf39K5BvgHB7b4rHXO/cfP+ArQHK+AKcmHZpYOrAdS8YokAJxz7zjnNvkZJwHLgQGH+Z6Lkwr80//5T8IrKs8ys2bAmcAd/u9VKvBkkWNtcs497ZzL8X/G5VEPSCuybhdQv9DjaLzf9UZ4p373lvM1RKqF4voMiEjobAOamFlUfoHmnBsMYGYb8P9gMrOBeC0ix+C1VsUA7xQ51vpCy2uBHmXNULg4BPbifXGW5NcStm0DXGJm5xR6Phr42l/+FjgR2OAv78ArMDP9x2VV9PVjC39+hTQF6gCzCjVaGRDpL7cCPi/H6+Zrgve+1hZatxavheqgjM65vf7rl/aZnu+cm2JmJwBv+K+xE8DMrgF+CyT629bzny/Ood5zcTY65wp31l8LHI3384wGNhc6VgQH/p4VXi6v3UBckXVxQHqhxx3wiu8BzrmsI3gtkSpNLWcilesnvOLkvENs9wbeKbxWzrl4vL5GRU+TtSq03BrYVFEhy2g98KpzrkGhW13n3OP+8/nF2fH+8rd4xdkJlK84K6utwD6ge6E88X7H+/y87UvY96CRhUWOm41XvORrDWw80sDOuW/xTiU+AWBmbYD/ALcAjf1TlwvZ/7MvmvNQ77k4LYqccs3/3VmP97vZpNCx4pxzhU87lvY5HcoyIMrMOhZa14sD+6ctwTtd/IWZldaaK1KtqTgTqUTOuZ14/Yj+bWYX+32wIsysN1C30Kb1ge3OuQy/r1RxUw780czqmDfa7TpgUojjF/UacI6ZDTezSDOLNW/AQUv/+R/xTpcOAGY45xbhFTgDge8qOoxzLg+vsHnSzBLA6xdnZsP9TcYD15nZMP8zb2FmXfznUvD6kxV33Fy807mP+j+vNngtW69VUPR/AqeaWS+83wEHbPHzX4fXepovBWhpZrXK+J6LkwDcZmbRZnYJ0BXv9PRm4EvgH2YW539G7f3WvTIxTyz+qGP/dyLGz7oHeB942MzqmtlxeH+kvFr4GM65N4H7gSlmVlIxLVKtqTgTqWTOub/hfbnfg/dlmwK8APwer6ABuAnvSywdr3P328Uc6lu8juRTgSecc1+GOPoBnHPr8b5c78crJtbjdbaP8J/fA8wGFhU6RfUTXv+s1BDF+j3eZ/KzmaUBU/D70znnZuAVsU/i9XX6lv2tYU8BF/ujLYubL+1WvEEaq4BpeC2bL1VEYOfcFryBCg865xbj9Wf7Ce/3ogdeX718X+G1NP1qZlsP9Z5LMB1vIMlWvE79Fxfq/3cNXmG1GO809Lt4/efKqg1eS15+a9g+vD5t+W4CauP1e3sTGOsX7Qdwzr2CN6r5KzNLLMfri1QLdmDXAxEJd/6X1Woguph+VyIlMrMRwCjn3JCgs4hIydRyJiIiIhJGVJyJiIiIhBGd1hQREREJI2o5ExEREQkjKs5EREREwkiVu0JAkyZNXGJiYtAxRERERA5p1qxZW51zTcuzT5UrzhITE0lOTg46hoiIiMghmdnaQ291IJ3WFBEREQkjKs5EREREwoiKMxEREZEwUuX6nBUnOzubDRs2kJGREXQUkbARGxtLy5YtiY6ODjqKiIiUQ7UozjZs2ED9+vVJTEzEzIKOIxI45xzbtm1jw4YNtG3bNug4IiJSDtXitGZGRgaNGzdWYSbiMzMaN26s1mQRkSqoWhRngAozkSL0b0JEpGqqNsVZ0OrVq3fQuueff56JEycGkKZk33zzDWeffXbQMURERKQE1aLPWbgaM2ZMhRwnJyeHqCj9qERERGoCtZyF0J/+9CeeeOIJli5dyoABAwrWr1mzhh49egAwa9YsTjjhBPr168fw4cPZvHkzACeeeCJ33HEHSUlJPPXUU4wYMYIxY8aQlJREp06d+PTTTwGvv911111Hjx496NOnD19//XWp60VERCS8VbvmmD9/sojFm9Iq9Jjdjo7joXO6H/b+Xbp0ISsri9WrV9O2bVsmTZrEZZddRnZ2NrfeeisfffQRTZs2ZdKkSTzwwAO89NJLAGRlZRVcqmrEiBGsWbOGGTNmsHLlSk466SRWrFjBs88+i5mxYMECli5dymmnncayZctKXC8iIiLhrdoVZ+Hq0ksvZdKkSdx7771MmjSJSZMm8csvv7Bw4UJOPfVUAHJzc2nevHnBPpdddtlBx4iIiKBjx460a9eOpUuXMm3aNG699VbAKwLbtGnDsmXLSlwvIgc7++nvAfj01qEBJxERqYbF2ZG0cIXSZZddxiWXXMKFF16ImdGxY0cWLFhA9+7d+emnn4rdp27dugc8Ljr6TqPxRCrGwo0V29ouInIk1OeskrRv357IyEgeeeSRghaxzp07s2XLloLiLDs7m0WLFpV4jHfeeYe8vDxWrlzJqlWr6Ny5M0OHDuX1118HYNmyZaxbt67U9SJysE9uGcIntwwJOoaICFANW86CsnfvXlq2bFnw+Le//e1B21x22WXcfffdrF69GoBatWrx7rvvctttt7Fr1y5ycnK444476N69+Na/1q1bM2DAANLS0nj++eeJjY3lpptuYuzYsfTo0YOoqCgmTJhATExMietF5GA9WsYHHUFEpIA554LOUC5JSUkuv5N8viVLltC1a9eAElWOESNGcPbZZ3PxxRcHHUWqkJrwb0NEJJyZ2SznXFJ59tFpTRGp8Z6cvIwnJ2vAjIiEB53WrCImTJgQdASRauupqcsBuPPUTgEnERFRcSYiwu3DOgYdQUSkQMiKMzOLBb4DYvzXedc591CRbUYAfwc2+quecc6NC1UmEZHiqMVMRMJJKFvOMoGTnXO7zSwamGZmXzjnfi6y3STn3C0hzCEiIiJSZYSsOHPeMNDd/sNo/1a1hoaKSI2wYMMuQFNqiEh4COloTTOLNLO5QCow2Tk3vZjNLjKz+Wb2rpm1CmWeUNqwYQPnnXceHTt2pH379tx+++1kZWUBMHfuXD7//POCbfMviF6RnnjiCbp06ULv3r3p378/EydOBGDUqFEsXrwYgMcee+yAfQYPHlyhGUSqqnOemcY5z0wLOoaICBDi4sw5l+uc6w20BAaY2TFFNvkESHTO9QQmA68UdxwzG21myWaWvGXLllBGPizOOS688ELOP/98li9fzrJly9i9ezcPPPAAcHBxdqRyc3MPePz8888zefJkZsyYwdy5c5k6dSr589eNGzeObt26AQcXZz/++GOFZRKpyo5pEccxLeKCjiEiAlTSPGfOuZ3A18DpRdZvc85l+g/HAf1K2P9F51yScy6padOmIc16OL766itiY2O57rrrAIiMjOTJJ5/kpZdeIi0tjQcffJBJkybRu3dvJk2aBMDixYs58cQTadeuHf/6178KjvXaa68xYMAAevfuzY033lhQiNWrV4+77rqLXr16HXQtzscee4znnnuOuDjvyyUuLo5rr70WgBNPPJHk5GTuvfde9u3bR+/evbnqqqsKjpnv73//O/3796dnz5489JA3bmPPnj2cddZZ9OrVi2OOOaYgu0h18+mtQ3XRcxEJG6EcrdkUyHbO7TSz2sCpwP8V2aa5c26z//BcYMkRv/AX98KvC474MAc4qgec8XiJTy9atIh+/Q6sK+Pi4mjdujVr1qzh4YcfJjk5mWeeeQbwTmsuXbqUr7/+mvT0dDp37szYsWNZsWIFkyZN4ocffiA6OpqbbrqJ119/nWuuuYY9e/YwcOBA/vGPfxzwOmlpaaSnp9OuXbtS38Ljjz/OM888w9y5cw967ssvv2T58uXMmDED5xznnnsu3333HVu2bOHoo4/ms88+A2DXrl1l+bRERETkCIRytGZz4BUzi8RroXvbOfepmT0MJDvnPgZuM7NzgRxgOzAihHnCyllnnUVMTAwxMTEkJCSQkpLC1KlTmTVrFv379wdg3759JCQkAF5r3EUXXRSSLF9++SVffvklffr0AWD37t0sX76coUOHctddd/H73/+es88+m6FD1bIgIiISaqEcrTkf6FPM+gcLLd8H3FehL1xKC1eodOvWjXffffeAdWlpaaxbt44OHTowe/bsg/YpfBHyyMhIcnJycM5x7bXX8te//vWg7WNjY4mMjDxofVxcHPXq1WPVqlWHbD0riXOO++67jxtvvPGg52bPns3nn3/OH/7wB4YNG8aDDz5YzBFEqrYBj04BYMYDpwScRERE19asEMOGDWPv3r0FIyRzc3O56667GDFiBHXq1KF+/fqkp6eX6TjvvvsuqampAGzfvp21a9cecr/77ruPm2++mbS0NMBr+crPUlh0dDTZ2dkHrR8+fDgvvfQSu3d7M59s3LiR1NRUNm3aRJ06dfjNb37D3XffXWyRKVIdpKZnkpqeeegNRUQqgS7fVAHMjA8++ICbbrqJRx55hLy8PM4888yC0ZEnnXQSjz/+OL179+a++0puKOzWrRt/+ctfOO2008jLyyM6Oppnn32WNm3alPr6Y8eOZffu3fTv35/o6Giio6O56667Dtpu9OjR9OzZk759+/L6668XrD/ttNNYsmQJgwYNAryBAq+99horVqzg7rvvJiIigujoaJ577rnD+XhEwt70+4cFHUFEpIDlT7lQVSQlJbnk5OQD1i1ZsoSuXbsGlEgkfOnfhohIsMxslnMuqTz76LSmiIiIVEm5eY5NO/exa182eXlVq7GpNDqtKSI13n3vzwfgrxf2DDiJiJRmX1Yuc9bvIHnNDmau2c6cdTvZnZkDQIRBXO1o4gvd8h83KLI+vnY08XX2L9eLicLMAn53+6k4E5Ea780Z6wEVZyLhZtvuTJLX7mDm6u3MXLuDRRt3kZPnMIPOzepzfp+j6XJUHPuyctm1L/uA28592WzYsa/gcW4pLWuREUZcbBQN6tQqUuBF0aB2rYOLvUKFXZ1akRVe2Kk4E5Ea77ELegQdQaTGc86xdtteZq7Z7rWMrd3Oqi17AKgVFUHvlg0YfXw7+ic2om/rhsTXiS7Xsfdk5bJzb1ZBsZZWuJDbe2Bht2tvFuu27WGnv11pZ0yjIuyglrj4Qi12h0PFmYjUeFcObB10BJEaJyc3j8Wb05i5ZgfJa7Yzc80Otu72prSJrx1NUpuGXNKvFf0TG9KjZTwxUQfP9VlWZka9mCjqxUTRsmH59s3Lc+zOymHX3v1F3c4irXReQefdb9udxaote7xtMw6evqosVJyJiIhIyO3JzGHu+p0FLWOz1+1gb5Z3/eiWDWsztGMTkhIb0j+xER2a1iMiIjz6gEVEGHGx0cTFRtOqnPvm5TkiD2NufI3WrCApKSlceeWVtGvXjn79+jFo0CA++OCDSnv9CRMmcMstt1Ta61WW/Au3FzVq1CgWL15c5uO8/vrr9OzZkx49ejB48GDmzZtX8Nx///tfOnfuTIcOHXj88f3/iq666io6d+7MMcccw/XXX18wga9zjttuu40OHTrQs2fPEifnzczM5LLLLqNDhw4MHDiQNWvWAJCVlcV1111Hjx496NWrF998802x+2/bto2TTjqJevXqHfCz3bt3L2eddRZdunShe/fu3HvvvWX+HKR4UxanMGVxStAxRKqVLemZfLFgMw9/sphzn5lGzz9/yVXjpvPU1OVs25PFxf1a8vQVffjpvpOZ9vuTefKy3lw1sA2dmtUPm8LsSB3u+1DLWQVwznH++edz7bXX8sYbbwCwdu1aPv7444CTVV/jxo0r1/Zt27bl22+/pWHDhnzxxReMHj2a6dOnk5uby80338zkyZNp2bIl/fv359xzz6Vbt25cddVVvPbaawBceeWVjBs3jrFjx/LFF1+wfPlyli9fzvTp0xk7dizTp08/6DXHjx9Pw4YNWbFiBW+99Ra///3vmTRpEv/5z38AWLBgAampqZxxxhnMnDmTiIgD/1aKjY3lkUceYeHChSxcuPCA5373u99x0kknkZWVxbBhw/jiiy8444wzyvWZyH6jJnp/AKx5/KyAk4hUTc45Vm/dUzCKMnntDlZv9fqLxURF0LtVA8ae0J6kxIb0bdOQuNjD64tVU6jlrAJ89dVX1KpVizFjxhSsa9OmDbfeeisAa9asYejQofTt25e+ffvy448/AvDNN99wwgkncN5559GuXTvuvfdeXn/9dQYMGECPHj1YuXIlAFu2bOGiiy6if//+9O/fnx9++KHYHJs2beL000+nY8eO3HPPPQXrx44dS1JSEt27d+ehhx4qWJ+YmMg999xDjx49GDBgACtWrABgxIgRjBkzhqSkJDp16sSnn34KeJeluvvuu+nfvz89e/bkhRdeKHgfJ554IhdffDFdunThqquuIn9y48TERB566CH69u1Ljx49WLp0KQB79uzh+uuvZ8CAAfTp04ePPvoI8C72fvnll9O1a1cuuOAC9u3bV+x7LdyiVq9ePR544AF69erFscceS0rKwS0ggwcPpmFDr6PBsccey4YNGwCYMWMGHTp0oF27dtSqVYvLL7+8IMuZZ56JmWFmDBgwoGCfjz76iGuuuQYz49hjj2Xnzp1s3rz5oNf86KOPuPbaawG4+OKLmTp1Ks45Fi9ezMknnwxAQkICDRo0KLZ1sG7dugwZMoTY2NgD1tepU4eTTjoJgFq1atG3b9+CbHJ4hnVJYFiXhKBjiFQZ2bl5zF2/k3Hfr+LGV5NJ+ssUTv7Ht9zz3nymLEmhQ0I97j+zC+/fNJgFfxrOpBsH8bvhnTmxc4IKszKoli1nifd+Bhz4V/DICTOZujSVcdckcUq3ZgC8MX0d93+wgCsGtCoYQp+SlsHAx6aSUD+mzBdBXrRoEX379i3x+YSEBCZPnkxsbCzLly/niiuuKPgynjdvHkuWLKFRo0a0a9eOUaNGMWPGDJ566imefvpp/vnPf3L77bdz5513MmTIENatW8fw4cNZsmTJQa8zd+5c5syZQ0xMDJ07d+bWW2+lVatWPProozRq1Ijc3FyGDRvG/Pnz6dnTe7/x8fEsWLCAiRMncscddxQUYmvWrGHGjBmsXLmSk046iRUrVjBx4kTi4+OZOXMmmZmZHHfccZx22mkAzJkzh0WLFnH00Udz3HHH8cMPPzBkyBAAmjRpwuzZs/n3v//NE088wbhx43j00Uc5+eSTeemll9i5cycDBgzglFNO4YUXXqBOnTosWbKE+fPnl/q55tuzZw/HHnssjz76KPfccw//+c9/+MMf/lDi9uPHjy9oZdq4cSOtWu3vRdCyZcuDWsGys7N59dVXeeqpp0rcZ+PGjTRv3vyA/QpvFxUVRXx8PNu2baNXr158/PHHXHHFFaxfv55Zs2axfv16BgwYcMj3WtTOnTv55JNPuP3228u9r+w3fkT/oCOIhLXdmTnMWbejoPP+nHU72Zft9Rdr3agOJ3RuSv/ERvRPbEi7JuHTX6yqqpbFWdBuvvlmpk2bRq1atZg5cybZ2dnccsstzJ07l8jISJYtW1awbf/+/Qu+1Nu3b19Q7PTo0YOvv/4agClTphzQvyotLY3du3dTr169A1532LBhxMfHA951OteuXUurVq14++23efHFF8nJyWHz5s0sXry4oDi74oorCu7vvPPOgmNdeumlRERE0LFjR9q1a8fSpUv58ssvmT9/Pu+++y4Au3btYvny5dSqVYsBAwbQsmVLAHr37s2aNWsKirMLL7wQgH79+vH+++8D8OWXX/Lxxx/zxBNPAJCRkcG6dev47rvvuO222wDo2bNnQc7S1KpVi7PPPrvgNSZPnlzitl9//TXjx49n2rRphzxuvptuuonjjz+eoUOHlnmf0lx//fUsWbKEpKQk2rRpw+DBg4mMLP8opJycHK644gpuu+022rVrVyHZREQKW7VlN799ex7zN+wkz3kTvXY7Oo7L+reif2IjkhIb0iwu9tAHknKplsVZcf1GivvL+MqBrQ8aQt8sLrbc/U66d+/Oe++9V/D42WefZevWrSQleZfSevLJJ2nWrBnz5s0jLy/vgNNUMTExBcsREREFjyMiIsjJ8WY9zsvL4+effz7o9FZRhY8VGRlJTk4Oq1ev5oknnmDmzJk0bNiQESNGkJGRUbBd4YnzSlrOf+yc4+mnn2b48OEHPPfNN98U+9pFcxVe75zjvffeo3PnzqW+p7KIjo4uyFv0tQubP38+o0aN4osvvqBx48YAtGjRgvXr1xdss2HDBlq0aFHw+M9//jNbtmwpOIVb2j4PPPAAn33mtdrOnTu3YLuWLVuSk5PDrl27aNy4MWbGk08+WbD/4MGD6dSpEx988AF//vOfAa9PXf7vT0lGjx5Nx44dueOOO8ryMYmIlMu+rFzGvjab1PQMbjmpA/3bNqJP64bUi6mWpUNYUZ+zCnDyySeTkZHBc889V7Bu7969Bcu7du2iefPmRERE8Oqrr5Kbm1uu45922mk8/fTTBY/nzp1b5n3T0tKoW7cu8fHxpKSk8MUXXxzw/KRJkwruBw0aVLD+nXfeIS8vj5UrV7Jq1So6d+7M8OHDee655wpGLS5btow9e/aU673kGz58OE8//XRB37Q5c+YAcPzxxxcMqli4cCHz588/rOMXtW7dOi688EJeffVVOnXqVLC+f//+LF++nNWrV5OVlcVbb73FueeeC3gF0v/+9z/efPPNAzrrn3vuuUycOBHnHD///DPx8fE0b96cRx99lLlz5xb8fM4991xeeeUVAN59911OPvlkzIy9e/cWfG6TJ08mKiqKbt26ccEFFxTsf6jC7A9/+AO7du3in//8Z4V8PjVd4r2fFXSHEBHvD+gHPlzAstR0nrq8D789rTNDOzZVYVZJ9ClXADPjww8/5M477+Rvf/sbTZs2pW7duvzf//0f4J0Wu+iii5g4cSKnn346devWLdfx//Wvf3HzzTfTs2dPcnJyOP7443n++efLtG+vXr3o06cPXbp0oVWrVhx33HEHPL9jxw569uxJTEwMb775ZsH61q1bM2DAANLS0nj++eeJjY1l1KhRrFmzhr59++Kco2nTpnz44Yflei/5/vjHP3LHHXfQs2dP8vLyaNu2LZ9++iljx47luuuuo2vXrnTt2pV+/fod1vGLevjhh9m2bRs33XQT4PUBS05OJioqimeeeYbhw4eTm5vL9ddfT/fu3QEYM2YMbdq0KShaL7zwQh588EHOPPNMPv/8czp06ECdOnV4+eWXi33NkSNHcvXVV9OhQwcaNWrEW2+9BUBqairDhw8nIiKCFi1a8Oqrr5aYOzExkbS0NLKysvjwww/58ssviYuL49FHH6VLly4FffJuueUWRo0aVSGflYjIpJnreX/2Rm4f1pHjOzUNOk6NY/ktF1VFUlKSKzqybcmSJXTt2jWgRFVXYmIiycnJNGnS5ID1I0aM4Oyzz+biiy8OKJlUFP3bEJHyWrhxFxc+9yMD2zZiwnUDiFTn/iNiZrOcc6WfDilCpzVFREQEgF37srn5jdk0qlOLf17WW4VZQHRaswbLn7G+qAkTJlRqDhERCZ5zjrvfmcfGHfuYdOOxNK4Xc+idJCTUciYiNd7ICTMZOWFm0DFEAjXu+9V8uTiFe8/oQr82jYKOU6NVm5Yz59xB0z+I1GRVrT9pkKYuTQ06gkigZq7ZzuP/Xcrp3Y9i5JC2Qcep8apFcRYbG8u2bdsK5pASqemcc2zbtu2Qc+OJZ9w15eqrK1KtbN2dyS1vzKZVw9r87ZKe+h4NA9WiOGvZsiUbNmxgy5YtQUcRCRuxsbEFV22Q0uVf0k2kpsnNc9zx1lx27s3mpZv667qXYaJaFGfR0dG0batmWBERkfJ4aupypq3Yyt8u6kn3o+ODjiO+kA0IMLNYM5thZvPMbJGZ/bmYbWLMbJKZrTCz6WaWGKo8IiIleWP6Ot6Yvi7oGCKV6ttlW3j6q+Vc3K8ll/ZvFXQcKSSULWeZwMnOud1mFg1MM7MvnHM/F9pmJLDDOdfBzC4H/g+4LISZREQOcv8HCwAOutauSHW1aec+7nhrDp2b1eeR844JOo4UEbLizHlDxXb7D6P9W9HhY+cBf/KX3wWeMTNzGmYmIpXoigFqNZCaIysnj5vfmE12ruPfV/Wldq3IoCNJESHtc2ZmkcAsoAPwrHNuepFNWgDrAZxzOWa2C2gMbC1ynNHAaPCu+SgiUpH+emHPoCOIVJq/frGEOet28u+r+tKuab2g40gxQjoJrXMu1znXG2gJDDCzw2o7dc696JxLcs4lNW2qC7CKiIgcjs8XbOblH9Zw3XGJnNmjedBxpASVcoUA59xO4Gvg9CJPbQRaAZhZFBAPbKuMTCIi+VLSMkhJywg6hkhIrdqym3venU+f1g2474yuQceRUoRytGZTM2vgL9cGTgWWFtnsY+Baf/li4Cv1NxORyjbwsakMfGxq0DFEQmZfVi43vT6b6Ejj2Sv7UitKV28MZ6Hsc9YceMXvdxYBvO2c+9TMHgaSnXMfA+OBV81sBbAduDyEeUREipVQXxd4lurtwY8W8ktKOi+P6M/RDWoHHUcOIZSjNecDfYpZ/2Ch5QzgklBlEBEpixkPnBJ0BJGQeXvmet6ZtYHbhnXkxM4JQceRMlC7poiISDW1eFMaf/xoIUM6NOH2YR2DjiNlpOJMRESkGkrLyOam12fRoE40/7y8N5ERuqB5VVEtrq0pInIkzn76ewA+vXVowElEKoZzjnvemc/6HfuYNPpYmtRTv8qqRMWZiNR4CzemBR1BpEKNn7aa/y76lT+c1ZWkxEZBx5FyUnEmIjXeJ7cMCTqCSIWZtXY7j3+xlOHdmzFySNug48hhUHEmIjVej5bxQUcQqRDbdmdy8+tzaNGwNn+7uBdm6mdWFak4ExERqQZy8xx3TJrL9r1ZfHDTYOJrRwcdSQ6TRmuKSI335ORlPDl5WdAxRI7I018t5/vlW3n43O50P1qtwVWZijMRqfGemrqcp6YuDzqGyGH7btkWnpq6nIv6tuSy/q2CjiNHSKc1RaTG0+ScUpVt3rWPOybNpXOz+vzl/GPUz6waUHEmIjXenad2CjqCyGHJzs3j5tdnk5mdy7NX9aV2rcigI0kFUHEmIiJSRT3+xVJmr9vJM1f2oX3TekHHkQqiPmciUuMt2LCLBRt2BR1DpFy+WLCZ8dNWM2JwImf3PDroOFKB1HImIjXeOc9MA2DN42cFnESkbFZv3cM9786nd6sG3H9m16DjSAVTcSYiNd4xLeKCjiBSZhnZuYx9bRaRkcazV/WlVpROglU3Ks5EpMbTBc+lKnnwo4X8kpLOyyP606JB7aDjSAio3BYREaki3k5ez9vJG7jlpA6c2Dkh6DgSIirOREREqoDFm9L444cLGdy+MXecoulfqjMVZyJS4w14dAoDHp0SdAyREqVnZHPzG7NpUCeaf13Rh8gITTRbnanPmYjUeKnpmUFHECmRc47fvzefddv38tboY2lSLyboSBJiKs5EpMabfv+woCOIlOjlH9bw+YJfuf/MLvRPbBR0HKkEKs5EpMZrFhcbdASRYs1au4PHPl/Cad2accPQdkHHkUqiPmciIiJhaPueLG55YzZHN6jN3y/ppQua1yBqORORGu++9+cD8NcLewacRMSTm+e4/a05bNuTxftjBxNfOzroSFKJSmw5M7N0M0sr6XaoA5tZKzP72swWm9kiM7u9mG1ONLNdZjbXvz14pG9IRKS83pyxnjdnrA86hkiBZ75awffLt/Lnc7tzTIv4oONIJSux5cw5Vx/AzB4BNgOvAgZcBTQvw7FzgLucc7PNrD4wy8wmO+cWF9nue+fc2YeVXkSkAjx2QY+gI4gUmLZ8K/+cuowL+7Tg8v6tgo4jASjLac1znXO9Cj1+zszmAaW2cjnnNuMVdTjn0s1sCdACKFqciYgE6sqBrYOOIALA5l37uO2tOXRMqMdfLjhG/cxqqLIMCNhjZleZWaSZRZjZVcCe8ryImSUCfYDpxTw9yMzmmdkXZta9PMcVERGpLrJz87jljTlkZufy3G/6UaeWuoXXVGUpzq4ELgVS/Nsl/royMbN6wHvAHc65on3VZgNt/Ja5p4EPSzjGaDNLNrPkLVu2lPWlRUTKZMriFKYsTgk6htRwf/vvUmat3cHjF/WkfdN6QceRAJlzLnQHN4sGPgX+55z7f2XYfg2Q5JzbWtI2SUlJLjk5ueJCikiNl3jvZwCsefysgJNITfXfhZsZ89psrh3Uhj+fd0zQcaQCmdks51xSefY5ZJupmTUFbgASC2/vnLv+EPsZMB5YUlJhZmZHASnOOWdmA/Ba8raVOb2ISAUY1iUh6AhSg63Zuoe735lPr1YNuP+srkHHkTBQlhPaHwHfA1OA3HIc+zjgamCBmc31190PtAZwzj0PXAyMNbMcYB9wuQtlU56ISDHGj+gfdASpoXZn5nDT67OJiDCevbIPMVGRQUeSMFCW4qyOc+735T2wc24a3tQbpW3zDPBMeY8tIiJS1e3JzOG6l2fwS0o6465NomXDOkFHkjBRlgEBn5rZmSFPIiIiUkN4hdlMZq/byb8u78NJnXVqXfYrS3F2O16BluFfHSC9LFcIEBGpKhLv/axgUIBIqO3NyuG6CTOZtW4H/7ysN2f1LMu87lKTHPK0Zv6VAkREROTI7M3K4foJM0les51/Xt6Hc3odHXQkCUNlGa2Zf8mmts65R8ysFdDcOTcj5OlERCqBptCQyrAvK5eRE5KZsXo7T17Wm3NVmEkJynJa89/AIPZPPLsbeDZkiURERKqZfVm5jHxlJtNXb+P/Xdqb83q3CDqShLGyjNYc6Jzra2ZzAJxzO8ysVohziYiIVAv7snIZNXEmP63axv+7tBfn91FhJqUrS8tZtplFAg4KJqXNC2kqEZFKNHLCTEZOmBl0DKmGMrJzuWFiMj+u3MY/LunFBX1aBh1JqoCytJz9C/gASDCzR/Emjv1DSFOJiFSiqUtTg44g1VB+YfbDyq38/eJeXNhXhZmUTanFmZlFAKuBe4BheJPKnu+cW1IJ2UREKsW4a8p12TuRQ8rIzmX0q7OYtmIrf7uoJxf3U2EmZVdqceacyzOzZ51zfYCllZRJRKRSndKtWdARpBrJyM7lxldn8f3yLfzfhT25JKlV0JGkiilLn7OpZnaRP6WGiIiIlCAzJ5cxr83i22VbePzCHlzaX4WZlF+JxZmZPeYv3gi8A2TqCgEiUh29MX0db0xfF3QMqeIyc3IZ8+osvvllC3+9sAeX9W8ddCSpoko7rXk6cL+uECAi1d39HywA4MqB+jKVw5OZk8vY12bz9S9beOyCHlwxQL9LcvhKK84izawh3iCAgzjntocmkohI5bpigE49yeHLysnj5tdn89XSVB694BgV+XLESivOugCzKL44c0C7kCQSEalkf72wZ9ARpIrKysnjptdnM2VJKo+cfwxXDWwTdCSpBkorzhb7ozRFRESkiKycPG5+YzZTlqTwyHndufpYFWZSMcoyWlNEpFpLScsgJS0j6BhShWTn5nHrm7OZvDiFP5/bnasHJQYdSaqR0lrOnqq0FCIiARr42FQA1jx+VsBJpCrIzs3j1jfm8L9FKfzpnG5cOzgx6EhSzZRYnDnnJlRiDhGRwCTUjwk6glQR2bl53PbmHP676FcePLsbI45rG3QkqYbKcm1NEZFqbcYDpwQdQaqA7Nw87nhrLl8s/JU/nt2N64eoMJPQUJ8zERGRQ8jJzeOOSXP5bMFm/nBWV0aqMJMQOmRxZmadzGyqmS30H/c0sz+EPpqIiEjwcnLzuPPteXw2fzMPnNmVUUM1k5SEVllazv4D3AdkAzjn5gOXhzKUiEhlOvvp7zn76e+DjiFhKCc3j9++PY9P5m3ivjO6cMPxKswk9MrS56yOc25Gkeue54Qoj4hIpVu4UZcLloPl5jnuemceH8/bxL1ndOHGE9oHHUlqiLIUZ1vNrD3eVQEws4uBzSFNJSJSiT65ZUjQESTM5OY57np7Lh/N3cQ9p3dmjAozqURlKc5uBl4EupjZRmA1cNWhdjKzVsBEoBleYfeic+6pItsY3nxqZwJ7gRHOudnlegciIkeoR8v4oCNIGMnNc9z9zjw+nLuJu4d35qYTOwQdSWqYshRnzjl3ipnVBSKcc+lmVpZhKjnAXc652WZWH5hlZpOdc4sLbXMG0NG/DQSe8+9FREQqXW6e4+535/H+nI387rRO3HySCjOpfGUZEPAegHNuj3Mu3V/37qF2cs5tzm8F8/dbArQostl5wETn+RloYGbNy5xeRKQCPDl5GU9OXhZ0DAlYbp7j9+/N5/3ZG/ntqZ245eSOQUeSGqrEljMz6wJ0B+LN7MJCT8UBseV5ETNLBPoA04s81QJYX+jxBn/dAX3azGw0MBqgdevW5XlpEZFDemrqcgDuPLVTwEkkKHl5jnvfm8+7szZw5ymduG2YCjMJTmmnNTsDZwMNgHMKrU8HbijrC5hZPbzWtzucc4c1JMo59yJevzeSkpLc4RxDRKQkt+uLuEbLy3Pc9/4C3pm1gduHdeT2U/T7IMEq7dqaHwEfmdkg59xPh3NwM4vGK8xed869X8wmG4FWhR639NeJiFQatZjVXHl5jvs/WMCk5PXcdnIH7lBhJmGgLAMC5pjZzXinOAtOZzrnri9tJ38k5nhgiXPu/5Ww2cfALWb2Ft5AgF3OOU3TISIiIZeX53jgwwW8NXM9t57cgTtP7USROT1FAlGWAQGvAkcBw4Fv8Vq30kvdw3MccDVwspnN9W9nmtkYMxvjb/M5sApYgXclgpvK+wZERI7Ugg27WLBhV9AxpBLl5Tn+8NFC3pyxnptPas9vVZhJGClLy1kH59wlZnaec+4VM3sDOOR1Tpxz04BSf9Odcw5vHjURkcCc88w0ANY8flbASaQyOOf440cLeWP6Osae2J7fndZZhZmElbIUZ9n+/U4zOwb4FUgIXSQRkcp1TIu4oCNIJXHO8eBHi3h9+jrGnNCee4arMJPwU5bi7EUzawj8Aa+PWD3gjyFNJSJSiT69dWjQEaQSOOd46ONFvPrzWm48vh2/P12FmYSnQxZnzrlx/uJ3QDsAM9NkYyIiUmU45/jzJ4uZ+NNaRh/fjnvP6KLCTMJWqcWZmQ3CmxT2O+dcqpn1BO4FhnLgFBgi1dqGHXv57aR57NyXRWx0JLFRkcRERxDj38dGRRLrPz7wPsLbPjqyYDkmKoKY6JK2jyQ60vSlIVKBnHM8/OliJvy4hlFD2nKfCjMJc6VdIeDveJPQzgV+b2b/A0YBfwVKnUZDpDrZtTebES/PJCUtg+PaNyEjJ5eM7Fx2Z+awdXcWmTm5ZGbnkZGdS2aOd5+Td/hzJUcYBxVzBxZ1RZ/zisM6MVG0bVKHjgn16ZBQj9joyAr8FKq3AY9OAWDGA6cEnEQqknOO5am7eWnaat6auZ7rj2vLA2d1VWEmYa+0lrOzgD7OuQy/z9l64Bjn3JpKSSYSBjJzchn9ajLrtu1l4sgBHNuucZn2y8nNKyjUit5nZOeRmbP/PrPQ42K3z8kjs9D9rn3ZZBaz3b7sXJxfE0YYtG5Uh47N6tOpWT06NatPx4T6tGtaV0VbMVLTM4OOIBUkKyePGau3M2VJClOXprB++z4AbhjalvvPVGEmVUNpxVmGcy4DwDm3w8yWqzCTmiQvz/G7d+YzffV2nrq8d5kLM4CoyAiiIiOoG1OWMTcVIysnjzXb9rAsJZ1lKbtZnpLOspR0vlqaSq7fkhdhkNi4Lh39gq1DgnffrmldYqJqbtE2/f5hQUeQI7Bzbxbf/LKFyUtS+O6XLaRn5hATFcFxHZow5oT2DOvSjKPiy3VJaJFAlfbN0c7MPi70uG3hx865c0MXSyR4f//yFz6Zt4nfn96F83q3CDrOIdWKiqBTs/p0alb/gPVZOXms3uoVbcv9wm1ZajpTluwv2iIjjDaN69ApwWtp6+gfp22TutSKKstc1VVbszh9cVc1q7bsZsqSFKYsSWXW2h3k5jma1IvhzB7NGdY1gSEdm1CnVuX9cSRSkUr7zT2vyON/hDKISDh57ee1PPfNSn5zbGvGnNAu6DhHpFZUBJ2Pqk/now4s2jJzclm1Jb9o282ylHR+SUnny8W/kt9lLirCSGxS1yvYEryCrWOzerRtUpfoyOpftEn4yMnNY9baHd7pyiWprNq6B4AuR9Vn7AntGdY1gV4tGxARodOWUvWVduHzbysziEi4mLokhQc/WsiwLgn86Zzu1baPSkxUJF2bx9G1+YETsGZk57Jyy25WpO4uOEW6eFMaXyz8taBPW1SE0bZJ3YJirZPft61N46pZtN33/nwA/nphz4CTSGFpGdl8+8sWpi5J4etftrBrXzbRkcax7Rpz7eBEhnVNoGXDOkHHFKlwavMVKWTe+p3c8sYcuh8dz9NX9iGqChYaRyo2OpLuR8fT/ej4A9ZnZOeyInU3y1P392lbsHEXny/cXFC0RUca7ZrUO6Bg69isPm0a1Qnrz/LNGesBFWfhYN22vQWd+aev2k5OnqNhnWiGdU3glK7NGNqxCfVjo4OOKRJSKs5EfOu372XkKzNpXK8W40ckqb9KEbHRkRzTIp5jWhxYtO3Lyt3fypbqnSKdt2Enn87fXLBNrcgIOh1VjxM7JYTl6afHLugRdIQaKzfPMXf9Tv90ZQrLUnYD0CGhHiOHtuWUrs3o27ohkWH0+yISaubc4c/HFISkpCSXnJwcdAypZnbsyeKi539k2+4s3hs7mA4J9YKOVOXtzcrxizavlW3O+p0HdNwe1iVBHbdrqD2ZOXy/fCtTlqTw9dJUtu3JIjLCGJDYqKCFLLFJ3aBjilQIM5vlnEsqzz6H/B/RzDoBdwNtCm/vnDu53AlFwlBGtjeX2YYd+3ht5EAVZhWkTq0oerZsQM+WDQrW5U95MGVJCp8v2Myk5PUFUx4M65qgKQ+qsU079zHVH13508ptZOXmERcbxYmdvSL9xE4JxNfR6UoRKEPLmZnNA54HZgG5+eudc7NCG614ajmTipSX57j1rTl8Nn8zz1zZh7N7Hh10pBojKyePmWu2+9Mh7J8stEeL+ILWk+5Hx1XKgIwpi1MAOKVbs5C/Vk2Rl+dYuGkXUxZ7BdnizWkAtGlch1O6NmNY1wT6JzaqkgNIRMrjcFrOylKczXLO9TuiZBVIxZlUpMc+X8KL363i/jO7MPr49kHHqbHyL7OTP03C7HU7cA6OiostKNQGtW8csqsbJN77GQBrHj8rJMevKTKyc/lhxdaCn2NqeiYRBv3aNGRY12ac0jWB9k3rVdsR0CLFCclpTeATM7sJ+AAouMaJc257OfOJhJVXflzDi9+t4tpBbbhhaNWey6yqM7OCCXRvOrEDW3dn8vXSVKYuSeWDORt5ffo6akdHMqRjE07t2oyTuiTQtH5Mhb3+sC4JFXasmmBvVg4paZmkpmWQkp5Jyq4Mpq/exrQVW8nIzqNeTBTHd2rCsC7ez6pR3VpBRxapUsrScra6mNXOORfIt5lazqQifLnoV258bRandG3G87/pp5FgYSwjO5efV21j6pJUpi5JYdOuDMygV8sGnNI1gVO6NaNzs/pqjakA+7JySU3PICUtk5S0DFLT/QIszVuXmp5Balom6Zk5B+3bokFtTumawLCuzRjYrlGNvhyYSGEhOa0ZblScyZGas24HV/znZzofFcdbNxxL7Vr6EqkqnHMs3pxWUKjN27AL2F8YnNKtGQPbNq4Rl5wqj4zsXFL94iq/8EpJz2BLWiYphdalZxxcdNWKiqBZXAzN6seSEBdDQv1YmsXF0qxg2buPqx2lAlmkGKHqcxYNjAWO91d9A7zgnMs+nJBHSsWZHIm12/Zw4b9/pG5MFO/fNJgm9Sru1JhUvtS0DKYu9Qq1mnhKLTPnwKKr4DRjWsYB63ftO/i/6+hIKyiumsXFklA/hoS4/YVX/rr42tEqukSOQKiKs3FANPCKv+pqINc5N+qwUh4hFWdyuLbvyeKi535k515vLrN2TTVlRnWyL8vrjD51aUmd0ZvRvmndYguNcBkQkJGdy8692ezad/Bt2+5MUgsVXinpGezcW3LR1bR+TEGR1Swu/3FsQStYgzoqukQqQ6gGBPR3zvUq9Pgrf3oNkSojIzuXUa/MZOPOfbx5w0AVZtVQ7VqRnNKtGad0a3bQNA6Pf7GUx79YSmLjOgwL8TQOGdm5pBUprIoWXPnP7yyyXVZOXonHjYywgtat1o3r0L9tQ5r5pxgTCp1ibFinVlhdfUFEyq8sLWezgUuccyv9x+2Ad51zfSsh30HUcibllZvnuPn12fxv8a/8+8q+nNGjedCRpJJt3LmPr0qbALVzAvG190+AmpWTV6hoytq/vDebXfty2OmvK64IyyylwAKoHxNFXO1o4mtH06COd59/iytmXcFzsdEqukSqoFC1nN0NfG1mqwDDu1LAdYeRTyQQj362hP8u+pU/nt1NhVkN1aJBba4elMjVgxIPunTQx/M2ERVhJDapy+6MHHbty2Zfdm6px6tbK5IGdWr5RVYUbZvULSii9q/ff2vg39ePjQrrC8CLSHg4ZHHmnJtqZh2Bzv6qX5xzmaXtA2BmLwFnA6nOuWOKef5E4CMgf6qO951zD5cxt0iZjJ+2mpd+WM11xyUyckjboONIGKgbE8XpxxzF6cccdcBFt1em7j6wtapOodasIq1bmtVeREKpxOLMzE52zn1lZhcWeaqDmeGce/8Qx54APANMLGWb751zZ5ctqkj5fLFgM3/5bDGndz+KP5zVLeg4EoYiI4x+bRry769XAPD3S3odYg8RkdArreXsBOAr4JxinnNAqcWZc+47M0s8/Ggih2/W2u3cMWkufVo14J+X99Yks1KqqUtTg44gIlKgxOLMOfeQv/iwc+6AqwSYWUWdHxrkj/zcBPzOObeogo4rNdiqLbsZ9UoyzeNjGXdt/5Bdj1Gqj3HXlKuvrohISJVlQMB7QNGRme8CR3ox9NlAG+fcbjM7E/gQ6FjchmY2GhgN0Lp16yN8WanOtu7OZMTLMzEzJlw3oFpPQCoV55RuzYKOICJSoLQ+Z12A7kB8kX5ncUDskb6wcy6t0PLnZvZvM2vinNtazLYvAi+CN5XGkb62VE/7snIZ9UoyqekZvHHDsSQ2qRt0JBERkXIrreWsM95oywYc2O8sHbjhSF/YzI4CUpxzzswGABHAtiM9rtRMuXmO296aw7wNO3n+N/3o27ph0JGkCnlj+joArhyolnkRCV5pfc4+Aj4ys0HOuZ/Ke2AzexM4EWhiZhuAh/AuA4Vz7nngYmCsmeUA+4DLXVW7CruEBeccD3+yiMmLU/jTOd0Y3v2ooCNJFXP/BwsAFWciEh7K0udsjJktcc7tBDCzhsA/nHPXl7aTc+6KQzz/DN5UGyJHZNz3q3nlp7XcMLQtI47TXGZSflcMaBV0BBGRAmUpznrmF2YAzrkdZtYndJFEyu6z+Zt59PMlnNWjOfed0TXoOFJF/fXCnkFHEBEpUJZpriP81jIAzKwRZSvqREJq5prt3Pn2XJLaNOQfl/bSdQdFRKRaKEuR9Q/gJzN7x398CfBo6CKJHNpKfy6zlg1r859rkjSXmRyRlLQMAJrFHfFAdBGRI1aWa2tONLNk4GR/1YXOucWhjSVSsi3pmYx4eQbRkcaEEQNoqLnM5AgNfGwqAGsePyvgJCIiZT892QjY45x72cyamlnbolcNkPCXlZNHhEFUFb5o896sHEa+MpMt6ZlMGj2I1o3rBB1JqoGE+jFBRxARKXDI4szMHgKS8OY9exlvOozXgONCG00qUmpaBpe+8BNbd2dxbLvGDO3YhOM6NKF907qYVY2+Wjm5edz6xhwWbtzFi1cn0atVg6AjSTUx44FTgo4gIlKgLC1nFwB98C63hHNuk5nVD2kqqVA792Zx9fgZbEnP5Kyezfl51XamLEkBoHl8LEM6NGGIX6w1qReeLQjOOf70ySKmLk3lkfO663I7IiJSbZWlOMvyZ/F3AGama+JUIXsycxjx8kxWb93DhOv6M7hDEwDWbdvL9yu28MOKrXy5OIV3Zm0AoGvzOIZ0aMyQjk0ZkNiI2rXCo6P9C9+t4rWf13HjCe24elBi0HFERERCxg41Kb+Z/Q7vguSnAn8FrgfecM49Hfp4B0tKSnLJyclBvHSVk5mTy8gJyfy0ahv/vqpviTPn5+Y5Fm7cxbQVW5m2fCuz1u4gKzePWpERJCU25LgOTRjasQndj44nMoDpKj6au5Hb35rLOb2O5qnLemvKDKlwZz/9PQCf3jo04CQiUt2Y2SznXFK59impODOzGOdcpr98KnAaYMD/nHOTjzTs4VJxVjY5uXnc8sYc/rvoV564pBcX92tZ5n33ZuUwc80Opi3fwvfLt7L013QAGtSJZnD7xgzp0JShHZvQqlHoO+P/vGob14yfQe/WDXh15ABiosKjJU+ql8R7PwM0WlNEKl5FF2eznXN9zexV59zVFZKwAjRJ7OrWL1sYNqfbwpFzjnvenc87szbw4NnduH7IkV3SaEt6Jj+u3Mr3y72WtV/9OaFaN6rDkI5NGNqhCYPaN6ZBnYqd0mJ5SjoXPfcjCXGxvDdmMPF1oiv0+CL5FmzYBUCPlvEBJxGR6qaii7OFwGPAI8DdRZ93zr1/OCGPVEzzju7ch15h/Ij+xMXqy7oo5xyPfraEcdNWc/uwjtx5aqcKP/7KLXuYtnwL01Zs5edV29mdmUOEQY8W8Qzp2IQhHZrSt02DI2rlSk3L4IJ//0hWbh7vjx1cKa10IiIiFa2ii7MhwFXApcDHRZ52h7rweah07N7LufP+Suej6jPx+gE0DtPRhUF55qvlPPHlMkYMTuShc7qFfJqM7Nw85q3f6bWqrdjK3PU7yc1z1I6OZEDbRgVTdnQ5qn6Zs+zJzOGyF39i1ZY9TBo9SK0ZIiJSZVVocVbooCOdc+OPKFkFSkpKcn9/7XPGvDaLlg1r89qogTSPrx10rLDw6k9r+ONHi7iwTwueuCSYa02mZ2Tz86rtBS1rK7fsAaBJvRiGdGjsDy5oylHxxV8mJyc3j1ETk/lu2RbGX9ufk7okVGZ8qaGenLwMoMJbmkVEQlKc+QceDCRSaOoN59zE8gasCPkDAqav2sbIV5KJrx3N66MGktikZs/w8dHcjdwxaS7DujTjud/0JTpMrgKwaec+pq3Yyg/+bevuLAA6JNTz5lfr0IRj2zemXkwUzjnu/2ABb85Yz2MX9ODKga0DTi81hQYEiEiohKrl7FWgPTAXyPVXO+fcbYcT8kgVHq25YMMurnlpOpEREbw2agBdjooLIlLgpi5JYfSrsxiQ2IiXr+sfthcBz8tzLP01nWkrtjBtxTZmrN5GRnYeURFGn9YNSIiL5bP5m7n5pPbcPbxL0HGlBlHLmYiESqiKsyVAN1eWJrZKUHQqjeUp6fxm/HQysvOYcF1/+rRuGGC6yvfzqm1c+9IMOh9VnzduOJZ6MWW9XGrwMrJzmb12hze/2oqtLNi4iwt6t+Afl/aqMpeUEhERKU2oirN3gNucc5uPJFxFKW6es/Xb93LVuOls3Z3JuGuSCmbBr+4WbtzF5S/+zFHxsbx94yAa1a3YqSwq257MHOrUilRhJiIi1cbhFGdl6ZjUBFhsZv8zs4/zb4cXMTRaNarDu2MG0bJhbUZMmMnkxSlBRwq5Fam7uealGcTXjubVkQOqfGEGUDcmSoWZBGLBhl0Fc52JiAStLOfA/hTqEBUhIS6WSaMHMeLlGYx5bRb/uKQX5/dpEXSskNi4cx9Xj59OhBmva7SqyBE755lpgAYEiEh4OGRx5pz7tjKCVISGdWvx+g3HMuqVmdz59lzSM3O4+tg2QceqUFt3Z3L1uOnszsxh0uhBNX6UqkhFOKZFzRxMJCLhqcTizMzSgeI6pBneaM2w/N+sXkwUE64bwM2vz+aPHy4kPSObm07sEHSsCpGWkc0142ewadc+Xhs5kG5Hh+WPQKTK0QXPRSSclNjnzDlX3zkXV8ytfrgWZvlioyN5/up+nNvraP723194/IulhMlg08O2LyuXkRNmsjw1nReuTiIpsVHQkURERCQEqs68C+UUHRnBk5f1pl5sFM9/u5L0jGweOe+YQGbNP1JZOXmMfX0WyWt38PQVfTihU9OgI4mIiEiIVNviDCAywnj0/GOoHxvFC9+uYndmDk9c0itsZs8vi9w8x13vzOObX7bw1wt7cHbPo4OOJFLtDHh0CgAzHjgl4CQiImWbSuOwmNlLZpZqZgtLeN7M7F9mtsLM5ptZ3xDl4L4zunL38M58NHcTY1+bTUZ27qF3DAPOOR78aCGfzNvEvWd04YoBupyRSCikpmeSmp4ZdAwRESCExRkwATi9lOfPADr6t9HAcyHMws0ndeCR87ozZUkK1708k92ZOaF8uQrx9//9wuvT1zH2xPaMOaF90HFEqq3p9w9j+v3Dgo4hIgKEsDhzzn0HbC9lk/OAic7zM9DAzJqHKg/A1YMSefKyXsxYs53fjJvOzr1ZoXy5I/LCtyv59zcruXJga+4Z3jnoOCLVWrO4WJrFxQYdQ0QECG3L2aG0ANYXerzBX3cQMxttZslmlrxly5YjetEL+rTk31f1ZfGmNC574WdS0zKO6Hih8NaMdfz1i6Wc3bM5j5x3jGbNFxERqUGqRM9459yLzrkk51xS06ZHPlJxePejePm6/qzfsZdLXviJ9dv3VkDKivHZ/M3c98ECTuzclP93aW8iq+DoUpGq5r7353Pf+/ODjiEiAgRbnG0EWhV63NJfVymO69CEV0cOZMeeLC55/idWpKZX1kuX6NtlW7hj0hz6tW7Ic1f1o1ZUlaidRaq8N2es580Z6w+9oYhIJQjy2/9j4Bp/1OaxwC7n3ObKDNCvTUMm3TiInDzHpS/8zMKNwV34eNba7Yx5dRYdE+ozfkR/ateKDCyLSE3z2AU9eOyCHkHHEBEBwEI1c76ZvQmcCDQBUoCHgGgA59zz5nWkegZvROde4DrnXPKhjpuUlOSSkw+5Wbms3rqH34ybTtq+bMaP6M+AtpU7+/6SzWlc9sJPNK4Xw9s3DqJp/ZhKfX0REREJDTOb5ZxLKtc+Ve2yRqEozgA27dzHb8ZPZ9POfTz/m36c2Dmhwl+jOGu27uHi538iOtJ4Z8wgWjasUymvKyIiIqF3OMWZOjX5jm5Qm7dvHES7JvW4YWIyn80P/RnWX3dlcNW46eQ5x6sjB6owEwnIlMUpTFmcEnQMERFAxdkBmtSL4c3Rx9KrZQNufXM2b88MXQfh7Xuy+M346ezal80r1w2gQ0K9kL2WiJRu1MRkRk2s+BZ5EZHDoeKsiPja0UwcOYDjOjThnvfmM37a6gp/jd2ZOVz38gzWbd/LuGuT6NEyvsJfQ0TKbliXBIZ1qZyuDCIih1KtL3x+uOrUimLctUnc8dZcHvl0MWn7srnjlI4VMhlsRnYuN7ySzMJNabzwm34c265xBSQWkSMxfkT/oCOIiBRQy1kJYqIiefqKPlzcryVPTV3OI58uIS/vyAZP5OTmceubc/hp1Tb+cUkvTunWrILSioiISHWhlrNSREVG8LeLelIvJoqXflhNekY2j1/U87Bm7c/Lc9zz3nwmL07h4fO6c36fYq9UJSIiIjWcirNDiIgwHjqnG3G1o/nX1OXsycrhyct6ExNV9klinXM8/Oli3p+9kbtO7cQ1gxJDF1hEyi3x3s8AWPP4WQEnERFRcVYmZsZvT+1EXGwUf/lsCbszZ/HCb/qVeRb/p6YuZ8KPaxg5pC23nNwhxGlFRESkKlNxVg6jhrajfmwU976/gGtems74Ef2Ji40udZ+Xf1jNP6cs5+J+LXngzK4VMqhARCqWWsxEJJxoQEA5Xda/NU9f0Ye563dyxYs/s3V3ZonbvjdrA3/+ZDHDuzfj8Qt7EHEYfdVERESkZlFxdhjO7nk0L16TxIrU3Vz6wk9s3rXvoG2+XPQr97w3n+M6NOapy/sQFamPWkRERA5NFcNhOqlzAhOvH0BqWiYXP/cTq7fuKXjux5VbueXNOfRoEc+LVycRG132wQMiUvlGTpjJyAkzg44hIgKoODsiA9s15s0bjmVfdi6XPP8TS39NY976ndzwSjKJjesw4br+1I1Rtz6RcDd1aSpTl6YGHUNEBABz7sgmVq1sSUlJLjk5vK6BtyI1navGTScjO48Ig3qxUbw7ZjDN4mKDjiYiZZB/0XNNDC0iFc3MZjnnksqzj1rOKkCHhPq8O2YwDepEExUZwWsjB6owE6lCTunWTIWZiIQNnXOrIK0a1eG/tx9PVm4e8bVLn15DREREpCQqzipQ7VqR1Ead/0WqmjemrwPgyoGtA04iIqLiTESE+z9YAKg4E5HwoOJMRGq8Kwa0CjqCiEgBFWciUuP99cKeQUcQESmg0ZoiIiIiYUTFmYjUeClpGaSkZQQdQ0QE0GlNEREGPjYVgDWPnxVwEhERFWciIiTUjwk6gohIgZCe1jSz083sFzNbYWb3FvP8CDPbYmZz/duoUOYRESnOjAdOYcYDpwQdQ0QECGHLmZlFAs8CpwIbgJlm9rFzbnGRTSc5524JVQ4RERGpRvJyITPdv6VB1l5vvRlgYPkbWqF1/v0B2xW+P9T2Rfcrx/EPQyhPaw4AVjjnVgGY2VvAeUDR4kxERESqO+cge9/+oiojzbvPTPPWZaTtf67g+fSDl7N2B/1OQi6UxVkLYH2hxxuAgcVsd5GZHQ8sA+50zq0vZhs5UrtTYfFHsG8H1G4IdRpDnUbefe1G3nJ07aBTigTi7Ke/B+DTW4cGnEQkTOXmFF80ZaZDxq4ij4sWXIUe5+Uc+rWi60JMfYiN8+5j4qB+c/+xfyv8XK263n7OAe7A5QPuKWZdcdtTzu0Pcfw/31DujzvoAQGfAG865zLN7EbgFeDkohuZ2WhgNEDr1rq8Spll7oaln8GCt2Hl1+ByS98+uo5frBVXvPmPiz4XXadQE69I1bRwY1rQEaQmysuDPVsgfRNkZ0BeNuRme6ftCpZzvFtutr/Of3xY2xZ+vtB9wbb564scPzcLcvYd+v1ERBUpnOIhvuWBhVTh54oWYPn3kUGXJhUtvIqzjUDha6K09NcVcM5tK/RwHPC34g7knHsReBEgKSnJFbeN+HJzYNXXMP9tWPopZO+F+NYw5A7ocSk0aue1nu3bDnu3wV7/ft92f7nQ451rvccZO0t+vciYQsVaowMLuQMKu4b7H8fUV0EnYeWTW4YEHUGqG+e8/zt3bYS0jbBrfaHlDd4tbZNXBB0xg8hoiIj2CqTIKG85MhoiIgstR/nP+9vWquOvi96/T8HzRe5jihRRxRVZUbH6v72ChLI4mwl0NLO2eEXZ5cCVhTcws+bOuc3+w3OBJSHMU305Bxtnw/xJsOh97y+x2AbQ8zLoeSm0OhYiCg3Mrd/Mu5VVbo73n8xBxdy2/QVd/uOURf7zOyho/i0qIvrgU6r5j+s0gbpNoW4TqJfgLdduVA3/kpJw0qNlfNARpKrJ2usVV7vW+wXXxgOX0zYe3DfKIiGuBcS3gJb9vfv4Vt4pu1p1SiiiihRUBxVfUV4BJtVKyL7xnHM5ZnYL8D8gEnjJObfIzB4Gkp1zHwO3mdm5QA6wHRgRqjzV0raVsOAdr5Vs+0qvFavz6V5R1uEUiKqguZsio7xiqW6Tsu+Tl+v1Qyi2mCvSUrfll/2Piz31al7xVrfp/sKtbkKh5aYHLqtlTkSORG42pG/2C64NkLbh4BawfdsP3q9ugncar2knaH+yX3y1hLiW3nK9ZiqkpEzMuap1ljApKcklJycHHSM4e7bCwve9VrKNyYBB26HeKctu50JsFW4ByMvzWuj2bPVa/4q9FXouY1fxx4mKLblwK664i4yu1LdZqrw8yMnwRjRl7/Xuc/Yd+PiA215/+73eF0pMXKHTyw33t0zqdHKpnpy8DIA7T+0UcBIJubw82Lu1hNOMfqvX7l/B5R24X2y8X2T5hVacX3jFt/SW446uuD+IpVoxs1nOuaTy7KNzRVVB1h745QuvIFsx1WtdatYDTn0YjrnY+4+iOoiI2H+Ks2kZviRzMgsVa0ULuq2wJxV2p3inWvds8Tq1Fie2wcFFXL2EIgVdU++0a17ugQXRAUVTkXUHbJO/XZF1RQuvnMO8vmNUrHeaIyu95G0iorwBHYULtqIFXP597Yb7l6NqHV6mKuSpqcsBFWeHJTsD1v8M6SleQVP4hiv02B38/EG3Q21T9Pniti9uXS7s3uK1gKVtOvj/gqja+wuu/Bav/NOP8a285Zh6gXy8UjOpOAtXuTmw+huY/w4s+QSy93h/tR13m9dK1qxb0AmDFxXj/+dZhuLUOW8od34Rtzu1mKJuq3eKdc204k9ZHI7oOt4UJVG1vfvo2t66WnW84q/wuqjY/dtH14Howo9L2Saq9v4+hXm5sG/n/tPEJd7v8AZ8bJrjrSutKKxVr5RCroRiLza+SrXS3T6sY9ARqg7nvC4VK6Z4tzXTyjaSr7ws4uAbVuixFb9NSc/XbQItkqBbodau/OKrdsMq9fsq1Z9Oa4YT57wvy/lvw8L3vJaf2Hjodr7Xj6z1oAM79kvo5GZ7feMKF257t3mtT4WLpQMKr8KFVO2qNXIpa2/xhdy+HbB3RwnP7aTEQR8WeWDrW+2GUK8pNO4ATTp5twZtNNCjqshIg9XfwcqpXkG2c523vlF7r39rh2HeckRpxVJZi6qIqvPvRqQMdFqzqtq+2u/YPwm2rYDIWtBpuFeQdTxN/RiCEBkN9Y/ybjVBLb81L75l2fcpPOhjXwkFXP79rg1eH8k9W/bvHxENjdtDk45+wdbZX+7o9Y+T4OTlwa/z/WJsKqyf7s1/VasetD0ejrsd2g+DRm2DTipSLak4C8qebd60F/Pfhg0zvHWJQ2HwbV7H/toNg80ncigRkfv7CJbVvp3eHyBbl3mnkLcuh9SlsPTzA0fq1j+6UNHWaf9y3NEhaVVZsMEbXFKjp9TYsxVWfuUVYyun7i+kj+oBg2/1irFWA2tEH0SRoKk4q0xZe+GXz71WshVTvL9EE7rBKX/yOvY3aHXIQ4hUabUbQMsk71ZYThbsWOMVbVuXeUXb1mXeHy+ZhUblRtctvmhr1M7rf3eYznlmGgBrHj/rsI9R5eRmw4aZXjG2Ygpsngc47zR0h2FeMdb+5PLNiSgiFULFWajl5cLqb70vmSWfeJMS1j8aBt3sdew/6pigE4oEL6qWN0K36Chd57zBG0WLtnU/e5cly2cRXh+2wgVb/q1u40O+/DEt4ir4DYWpnev2t4yt+tYbJGOR3oSoJz0AHU6G5r01F5dIwDQgIBSc8/4Kze/Yv/tXb/6pbud5/cjaHKeO/SJHKmuPN2qwcNG2dTlsW37g6NPajYop2jrWjAEJ2ftgzQ/7O/Jv9eZzI66l1zrWYRi0PcFr0RSRkDicAQFVrzjr1tYlv/aw3+/E73tS3HJBvxQ7sI/KQetLOcYhj1d0mf2jLbcu8zo8dxruXUKp4/AjOu0iImWUlwe71hUq2AoVb4UHJETW8kYYFpwabetfPqzJ/suJxcRVrZGDznl9+fKLsbU/eoVqZAwkDvELslO891uV3pdIFVYzirOjI13y6DCfDLD1YK8g63Ze+TpLi0ho7d2+f0BC4aJt++riLx0WEe0VaQUFW5NCjxsfeKvbJJhJe/ft9LpOrPBHVqZt8NY36eQVYu2HQZvB3mhcEal0NaM469PTJX/9qfcXIuDNQF3cMv6yK9vyIY/niuxH8ceLO7p80xGISOAG/+ULEtjGhyO6epf22btt/9x2hW/56zJ2lnywmHj/WrD5BVyTIo+LrCtv61xeHmyeu78j/4aZXmEZEwftTvCKsQ7DoEHrI/1YRKQC1Ix5ziJr6T8dEalQm3bnsYmG0LJf2XbIzfbmdiso4PILusLF3FbvWo2b53vLJV0+LL91rk5jb/BCQfFWZF3aZu905cqvvOMDHN0HhtzptZC1TAqv68SKyGGresWZiEgFm37/sPLtEBntXX+1XkLZtnfOG6l9QAG3tUiL3HZv3a8LvMfFtc7VbQodTvVaxtqd5F11QUSqHRVnIlLjNYsL8WAdM++qBzH1oWFi2fbJzfGvsuAXb7Hx0OwYjfQWqQFUnImIhKPIqPK1zolItaE/wUSkxrvv/fnc9/78oGOIiAAqzkREeHPGet6csT7oGCIigE5riojw2AU9go4gIlJAxZmI1HhXDtT0PCISPnRaU0RERCSMqDgTkRpvyuIUpixOCTqGiAig05oiIoyamAzAmsfPCjiJiIiKMxERhnXRXGIiEj5UnIlIjTd+RP+gI4iIFFCfMxEREZEwEtLizMxON7NfzGyFmd1bzPMxZjbJf366mSWGMo+IiIhIuAtZcWZmkcCzwBlAN+AKM+tWZLORwA7nXAfgSeD/QpVHRKQkifd+RuK9nwUdQ0QECG3L2QBghXNulXMuC3gLOK/INucBr/jL7wLDzMxCmElEREQkrIVyQEALoPDF6jYAA0vaxjmXY2a7gMbA1sIbmdloYDRA69aayVtEKpam0BCRcFIlBgQ45150ziU555KaNm0adBwRERGRkAllcbYRaFXocUt/XbHbmFkUEA9sC2EmERERkbAWyuJsJtDRzNqaWS3gcuDjItt8DFzrL18MfOWccyHMJCIiIhLWQtbnzO9DdgvwPyASeMk5t8jMHgaSnXMfA+OBV81sBbAdr4ATERERqbFCeoUA59znwOdF1j1YaDkDuCSUGURERESqkioxIEBERESkplBxJiIiIhJGVJyJiIiIhBEVZyIiIiJhRMWZiIiISBhRcSYiIiISRlSciYiIiIQRq2oT8ptZOvBL0DlKEQ/sCjpEKZTvyIRzvnDOBsp3pJTvyCjf4QvnbBD++To75+qXZ4eQTkIbIr8455KCDlESM3vROTc66BwlUb4jE875wjkbKN+RUr4jo3yHL5yzQZXIl1zefXRas+J9EnSAQ1C+IxPO+cI5GyjfkVK+I6N8hy+cs0H45yu3qnhaMzmcW85ERERE8h1O3VIVW85eDDqAiIiISBmVu26pcsWZcy5sijMzO93MfjGzFWZ2r7/OzOxRM1tmZkvM7LYwy/e9mc31b5vM7MMwyzfMzGb7+aaZWYcwy3eyn2+hmb1iZoH02zSzl8ws1cwWFlrXyMwmm9ly/75hENlKyXeJmS0yszwzC7T1u4R8fzezpWY238w+MLMGYZbvET/bXDP70syODqd8hZ67y8ycmTUJIpufobjP709mtrHQ/39nhks2f/2t/u/fIjP7WxDZSspnZpMKfW5rzGxumOXrbWY/+/mSzWxAUPmKc1h1i3NOt8O4AZHASqAdUAuYB3QDrgMmAhH+dgnhlK/INu8B14RTPmAZ0NXf5iZgQpjlWw908rd5GBgZUL7jgb7AwkLr/gbc6y/fC/xfENlKydcV6Ax8AyQFla2UfKcBUf7y/4Xh5xdXaPk24PlwyuevbwX8D1gLNAmnfMCfgN8FlekQ2U4CpgAx/uNAvjdK+9kWev4fwIPhlA/4EjjDXz4T+Cbon/OR3sK+5ayE1osJZra6UCXfO4BoA4AVzrlVzrks4C3gPGAs8LBzLg/AOZcaQLbS8gFgZnHAycCHwcQrMZ8D4vxt4oFNYZTvIiDLObfM32ayv67SOee+A7YXWX0e8Iq//ApwfmVmKqy4fM65Jc65sJgGp4R8XzrncvyHPwMtKz3Y/izF5Usr9LAu3r+VQJTw+wfwJHAPAWaDUvMFroRsY4HHnXOZ/jZBfW+U+tmZmQGXAm9WaqhCSsgXLt8bJbXs9TKzn8xsgZl94n//liqsizMziwSeBc7Aa7W4wsy6+U/f7Zzr7d/mBhCvBV4rSr4N/rr2wGV+0+oXZtYxgGxQcr585wNTi/yHX5lKyjcK+NzMNgBXA48HkA2Kz3cUEFXolNzFeC0F4aKZc26zv/wr0CzIMFXc9cAXQYcoyu8ysR64Cngw6DyFmdl5wEbn3Lygs5TiFv/U8EtBnvYvRidgqJlNN7Nvzax/0IFKMBRIcc4tDzpIEXcAf/f/bTwB3BdglgnA6UXWjcM7q9ED+AC4+1AHCevijEO0/oSpGCDDeSMz/gO8FHCeklxBgH/9lOJO4EznXEvgZeD/BZynMAdcDjxpZjOAdCA32EjFc177ftUaih0mzOwBIAd4PegsRTnnHnDOtcLLdkvQefKZWR3gfsKsYCziObw/nnsDm/FOz4WLKKARcCzeF/fbfitVuAnX742xwJ3+v407gfFBBSmhZa8T8J2/XKYzLuFenJXW+vOo/xfQk2YWU/nR2MiBrSYt/XUbgPf9dR8APSs5V76S8uF31B0AfBZArnzF5UsBejnnpvvrJgGDKzuYr9jPzzn3k3NuqHNuAN4/tmXF7h2MFDNrDuDfB3ZqpKoysxHA2cBVfoEbrl4noFPqJWgPtAXmmdkavH8vs83sqEBTFeKcS3HO5fpdTv6D939guNgAvO88M4A8ILABFcXxBz9diPf/cri5lv3fu+8QXj9bgEXsb1i6hDKccQn34qwk9wFdgP54f238PoAMM4GOZtbWzGrhtah8jNeH6yR/mxMI7su7pHzgnY771DmXEVA2KDlfvJl18rc5FVgSTvnMLAHA/4Pg98DzAeUrzsd4/0nh338UYJYqx8xOx+svda5zbm/QeYoq0kXiPGBpUFmKcs4tcM4lOOcSnXOJeMVGX+fcrwFHK5D/h4vvAuCgkaYB+hD/e8P//68WsDXIQMU4BVjqnNsQdJBibML7vgWvL3W4nXa9HrjJzGYB9YGsQ+4R9IiE0m7AIOB/hR7fB9xXZJsT8QqNIPKdiVd8rQQe8Nc1wGuRWgD8hNcSFNTnd1A+f/03wOlh8PMt7vO7wP/s5vk524VZvr/jFYy/AHcEmO1NvFMz2XhfhCOBxsBUvP+YpgCNwizfBf5yJl4r6f/CLN8KvJb6uf4tyNGQxeV7D6+gmI83I3qLcMpX5Pk1BDtas7jP71X//5b5eH/INA+jbLWA1/yf72zg5HD67Pz1E4AxQeU6xOc3BJjlf29MB/oFnDGRkke7dgJmHOoYYX2FAL8ZdRkwDO8000zgSmC7c26zf07+Sbw+XvcGl1REREQEzCwRr9HoGP9xgnMu1cwi8Ircb5xzpfZHD+sLnzvncszsFrx5cyKBl5xzi8zsKzNrChjeX7hjAowpIiIigpm9iXdGr4k/68BDQD0zu9nf5H28wW6lHyecW85EREREapqqOiBAREREpFpScSYiIiISRlSciYiIiISRsCzOzOwBM1vkTzI718wGBp1JREREpDKE3WhNMxuEN0N3X+dcpj+bfa2AY4mIiIhUinBsOWsObHXOZQI457Y65zaZWT//grCzzOx/hS5T842ZPeW3sC00s3C7bIOIiIhImYVjcfYl0MrMlpnZv83sBDOLBp4GLnbO9cO7mPijhfap45zrDdxE+F5oXEREROSQwu60pnNut5n1A4biXWtsEvAX4BhgsndRACLxLt+Q701/3+/MLM7MGjjndlZqcBEREZEKEHbFGYBzLhfvuorfmNkC4GZgkXNuUEm7HOKxiIiISJUQdqc1zayzmXUstKo33oWmm/qDBTCzaDPrXmiby/z1Q4BdzrldlZVXREREpCKFY8tZPeBpM2sA5AArgNHAi8C/zCweL/c/gUX+PhlmNgeIBq6v7MAiIiIiFaXKX1vTzL4BfuecSw46i4iIiMiRCrvTmiIiIiI1WZVvORMRERGpTgJvOTOzVmb2tZkt9i/ZdLu/vpGZTTaz5f59Q3/9Vf5lnRaY2Y9m1qvQsU43s1/MbIWZ3RvUexIRERE5XIG3nPkz/Td3zs02s/rALOB8YASw3Tn3uF9oNXTO/d7MBgNLnHM7zOwM4E/OuYFmFgksA04FNgAzgSucc4sDeFsiIiIihyXwljPn3Gbn3Gx/OR1v2owWwHnAK/5mr+AVbDjnfnTO7fDX/wy09JcHACucc6ucc1nAW/4xRERERKqMwIuzwswsEegDTAeaOefyrwLwK9CsmF1GAl/4yy2A9YWe2+CvExEREakywmaeMzOrB7wH3OGcS/Mv0wSAc86ZmSuy/Ul4xdmQSg0qIiIiEkJh0XLmX9j8PeB159z7/uoUvz9afr+01ELb9wTGAec557b5qzcCrQodtqW/TkRERKTKCLw4M6+JbDxeJ///V+ipj4Fr/eVrgY/87VsD7wNXO+eWFdp+JtDRzNqaWS3gcv8YIiIiIlVGOIzWHAJ8DywA8vzV9+P1O3sbaA2sBS51zm03s3HARf46gBznXJJ/rDPxLusUCbzknHu0st6HiIiISEUIvDgTERERkf0CP60pIiIiIvupOBMREREJIyrORERERMKIijMRERGRMKLiTERERCSMqDgTkRrDzP5kZr8r5fnzzaxbZWYSESlKxZmIyH7nAyrORCRQmudMRKo1M3sA7yojqcB6YBawCxgN1AJWAFcDvYFP/ed24U12DfAs0BTYC9zgnFtaifFFpAZScSYi1ZaZ9QMmAAOBKGA28Dzwcv51ec3sL0CKc+5pM5sAfOqce9d/biowxjm33MwGAn91zp1c+e9ERGqSqKADiIiE0FDgA+fcXgAzy7/e7jF+UdYAqAf8r+iOZlYPGAy8410CGICYUAcWEVFxJiI10QTgfOfcPDMbAZxYzDYRwE7nXO/KiyUiogEBIlK9fQecb2a1zaw+cI6/vj6w2cyigasKbZ/uP4dzLg1YbWaXAJinV+VFF5GaSsWZiFRbzrnZwCRgHvAFMNN/6o/AdOAHoHAH/7eAu81sjpm1xyvcRprZPGARcF5lZReRmksDAkRERETCiFrORERERMKIijMRERGRMKLiTERERCSMqDgTERERCSMqzkRERETCiIozERERkTCi4kxEREQkjKg4ExEREQkj/x98F80kK1CpCwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = plt.subplot(1, 1, 1)\n",
    "df.plot(ax=ax, figsize=(10,5))\n",
    "plt.vlines(x= pd.to_datetime(treat_date), \n",
    "           ymin=0, \n",
    "           ymax=1.1 * max(df.max()), linestyle=\":\", lw=2, label=\"Game happenned in {}\".format(treat_date))\n",
    "plt.ylabel(\"Infection Rate Trend\")\n",
    "plt.title(\"Gap in new Infection Rate per 10k\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acc9be18",
   "metadata": {},
   "source": [
    "# Matching Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "9899de8f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T22:59:59.028558Z",
     "start_time": "2022-04-14T22:59:58.710667Z"
    }
   },
   "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>areaidx</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>...</th>\n",
       "      <th>295</th>\n",
       "      <th>296</th>\n",
       "      <th>298</th>\n",
       "      <th>299</th>\n",
       "      <th>300</th>\n",
       "      <th>301</th>\n",
       "      <th>302</th>\n",
       "      <th>303</th>\n",
       "      <th>304</th>\n",
       "      <th>305</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>dateidx</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 rowspan=\"7\" valign=\"top\">newCasesRate</th>\n",
       "      <th>248</th>\n",
       "      <td>0.319707</td>\n",
       "      <td>0.778568</td>\n",
       "      <td>0.582971</td>\n",
       "      <td>0.253268</td>\n",
       "      <td>0.465541</td>\n",
       "      <td>1.146148</td>\n",
       "      <td>0.933147</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.038591</td>\n",
       "      <td>0.116550</td>\n",
       "      <td>...</td>\n",
       "      <td>0.098261</td>\n",
       "      <td>0.242186</td>\n",
       "      <td>0.294447</td>\n",
       "      <td>0.252346</td>\n",
       "      <td>0.374980</td>\n",
       "      <td>0.048145</td>\n",
       "      <td>0.421726</td>\n",
       "      <td>0.288871</td>\n",
       "      <td>0.121310</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>249</th>\n",
       "      <td>0.426276</td>\n",
       "      <td>0.778568</td>\n",
       "      <td>0.510100</td>\n",
       "      <td>0.253268</td>\n",
       "      <td>0.093108</td>\n",
       "      <td>1.480441</td>\n",
       "      <td>1.199760</td>\n",
       "      <td>0.578114</td>\n",
       "      <td>0.270139</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.131015</td>\n",
       "      <td>0.193749</td>\n",
       "      <td>0.425312</td>\n",
       "      <td>0.403753</td>\n",
       "      <td>0.156242</td>\n",
       "      <td>0.144434</td>\n",
       "      <td>0.512096</td>\n",
       "      <td>0.036109</td>\n",
       "      <td>0.242619</td>\n",
       "      <td>0.148232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>250</th>\n",
       "      <td>1.172258</td>\n",
       "      <td>0.637010</td>\n",
       "      <td>0.437229</td>\n",
       "      <td>0.557191</td>\n",
       "      <td>0.558649</td>\n",
       "      <td>1.910247</td>\n",
       "      <td>2.132907</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.540278</td>\n",
       "      <td>0.407924</td>\n",
       "      <td>...</td>\n",
       "      <td>0.327537</td>\n",
       "      <td>0.193749</td>\n",
       "      <td>0.752475</td>\n",
       "      <td>0.656098</td>\n",
       "      <td>0.499973</td>\n",
       "      <td>0.192579</td>\n",
       "      <td>0.662712</td>\n",
       "      <td>0.324980</td>\n",
       "      <td>0.636875</td>\n",
       "      <td>0.481753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>251</th>\n",
       "      <td>1.065689</td>\n",
       "      <td>1.274021</td>\n",
       "      <td>0.291486</td>\n",
       "      <td>0.557191</td>\n",
       "      <td>0.372433</td>\n",
       "      <td>1.910247</td>\n",
       "      <td>3.399320</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.115774</td>\n",
       "      <td>0.262237</td>\n",
       "      <td>...</td>\n",
       "      <td>0.458552</td>\n",
       "      <td>0.290623</td>\n",
       "      <td>0.392596</td>\n",
       "      <td>0.201876</td>\n",
       "      <td>0.468725</td>\n",
       "      <td>0.240724</td>\n",
       "      <td>0.632589</td>\n",
       "      <td>0.252762</td>\n",
       "      <td>0.363929</td>\n",
       "      <td>0.259405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>0.745982</td>\n",
       "      <td>0.566231</td>\n",
       "      <td>0.510100</td>\n",
       "      <td>0.810459</td>\n",
       "      <td>0.651757</td>\n",
       "      <td>1.766979</td>\n",
       "      <td>2.199560</td>\n",
       "      <td>0.361321</td>\n",
       "      <td>0.115774</td>\n",
       "      <td>0.611886</td>\n",
       "      <td>...</td>\n",
       "      <td>0.294783</td>\n",
       "      <td>0.193749</td>\n",
       "      <td>0.490745</td>\n",
       "      <td>0.353284</td>\n",
       "      <td>0.406228</td>\n",
       "      <td>0.096289</td>\n",
       "      <td>0.602466</td>\n",
       "      <td>0.397198</td>\n",
       "      <td>0.303274</td>\n",
       "      <td>0.222347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>253</th>\n",
       "      <td>1.278827</td>\n",
       "      <td>0.353895</td>\n",
       "      <td>0.145743</td>\n",
       "      <td>0.557191</td>\n",
       "      <td>0.372433</td>\n",
       "      <td>1.480441</td>\n",
       "      <td>1.732987</td>\n",
       "      <td>0.361321</td>\n",
       "      <td>0.038591</td>\n",
       "      <td>0.378787</td>\n",
       "      <td>...</td>\n",
       "      <td>0.196522</td>\n",
       "      <td>0.435935</td>\n",
       "      <td>0.556177</td>\n",
       "      <td>0.302815</td>\n",
       "      <td>0.437477</td>\n",
       "      <td>0.192579</td>\n",
       "      <td>0.421726</td>\n",
       "      <td>0.505525</td>\n",
       "      <td>0.667202</td>\n",
       "      <td>0.370579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>254</th>\n",
       "      <td>0.532845</td>\n",
       "      <td>0.353895</td>\n",
       "      <td>0.145743</td>\n",
       "      <td>0.354576</td>\n",
       "      <td>0.093108</td>\n",
       "      <td>2.005759</td>\n",
       "      <td>1.732987</td>\n",
       "      <td>1.373021</td>\n",
       "      <td>0.077183</td>\n",
       "      <td>0.145687</td>\n",
       "      <td>...</td>\n",
       "      <td>0.294783</td>\n",
       "      <td>0.193749</td>\n",
       "      <td>1.014205</td>\n",
       "      <td>0.100938</td>\n",
       "      <td>0.343732</td>\n",
       "      <td>0.288868</td>\n",
       "      <td>0.421726</td>\n",
       "      <td>0.541634</td>\n",
       "      <td>0.454911</td>\n",
       "      <td>0.148232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>People per sq. km</th>\n",
       "      <th>248</th>\n",
       "      <td>999.000000</td>\n",
       "      <td>2616.000000</td>\n",
       "      <td>560.000000</td>\n",
       "      <td>963.000000</td>\n",
       "      <td>541.000000</td>\n",
       "      <td>1163.000000</td>\n",
       "      <td>1092.000000</td>\n",
       "      <td>3999.000000</td>\n",
       "      <td>3629.000000</td>\n",
       "      <td>142.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>8702.000000</td>\n",
       "      <td>5490.000000</td>\n",
       "      <td>5411.000000</td>\n",
       "      <td>3450.000000</td>\n",
       "      <td>11045.000000</td>\n",
       "      <td>4706.000000</td>\n",
       "      <td>16427.000000</td>\n",
       "      <td>7137.000000</td>\n",
       "      <td>9624.000000</td>\n",
       "      <td>12167.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>old_dependency_ratio</th>\n",
       "      <th>248</th>\n",
       "      <td>0.313702</td>\n",
       "      <td>0.258597</td>\n",
       "      <td>0.378299</td>\n",
       "      <td>0.299317</td>\n",
       "      <td>0.333318</td>\n",
       "      <td>0.302608</td>\n",
       "      <td>0.229883</td>\n",
       "      <td>0.332554</td>\n",
       "      <td>0.231794</td>\n",
       "      <td>0.452757</td>\n",
       "      <td>...</td>\n",
       "      <td>0.135354</td>\n",
       "      <td>0.192277</td>\n",
       "      <td>0.194712</td>\n",
       "      <td>0.251617</td>\n",
       "      <td>0.118766</td>\n",
       "      <td>0.238610</td>\n",
       "      <td>0.088241</td>\n",
       "      <td>0.161879</td>\n",
       "      <td>0.133132</td>\n",
       "      <td>0.175793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>young_dependency_ratio</th>\n",
       "      <th>248</th>\n",
       "      <td>0.288634</td>\n",
       "      <td>0.315276</td>\n",
       "      <td>0.280907</td>\n",
       "      <td>0.299981</td>\n",
       "      <td>0.284279</td>\n",
       "      <td>0.279466</td>\n",
       "      <td>0.338812</td>\n",
       "      <td>0.285957</td>\n",
       "      <td>0.291240</td>\n",
       "      <td>0.261740</td>\n",
       "      <td>...</td>\n",
       "      <td>0.272450</td>\n",
       "      <td>0.298154</td>\n",
       "      <td>0.320712</td>\n",
       "      <td>0.307247</td>\n",
       "      <td>0.241648</td>\n",
       "      <td>0.314247</td>\n",
       "      <td>0.257283</td>\n",
       "      <td>0.307642</td>\n",
       "      <td>0.238689</td>\n",
       "      <td>0.223543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Average Score</th>\n",
       "      <th>248</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>40.500000</td>\n",
       "      <td>29.800000</td>\n",
       "      <td>25.800000</td>\n",
       "      <td>25.700000</td>\n",
       "      <td>18.900000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>40.600000</td>\n",
       "      <td>15.600000</td>\n",
       "      <td>...</td>\n",
       "      <td>26.700000</td>\n",
       "      <td>14.600000</td>\n",
       "      <td>17.200000</td>\n",
       "      <td>9.400000</td>\n",
       "      <td>25.800000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>27.900000</td>\n",
       "      <td>25.200000</td>\n",
       "      <td>16.600000</td>\n",
       "      <td>20.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>male_avg_age</th>\n",
       "      <th>248</th>\n",
       "      <td>40.157418</td>\n",
       "      <td>37.015974</td>\n",
       "      <td>42.037639</td>\n",
       "      <td>39.646713</td>\n",
       "      <td>41.155788</td>\n",
       "      <td>40.452154</td>\n",
       "      <td>36.626540</td>\n",
       "      <td>41.063122</td>\n",
       "      <td>37.117334</td>\n",
       "      <td>44.593287</td>\n",
       "      <td>...</td>\n",
       "      <td>35.124795</td>\n",
       "      <td>36.585779</td>\n",
       "      <td>35.694074</td>\n",
       "      <td>39.043512</td>\n",
       "      <td>34.824781</td>\n",
       "      <td>38.056436</td>\n",
       "      <td>32.614445</td>\n",
       "      <td>35.037225</td>\n",
       "      <td>34.995129</td>\n",
       "      <td>36.918846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>female_avg_age</th>\n",
       "      <th>248</th>\n",
       "      <td>42.190274</td>\n",
       "      <td>39.302367</td>\n",
       "      <td>44.115800</td>\n",
       "      <td>41.726862</td>\n",
       "      <td>42.778638</td>\n",
       "      <td>42.309614</td>\n",
       "      <td>37.817297</td>\n",
       "      <td>42.667796</td>\n",
       "      <td>38.678151</td>\n",
       "      <td>46.771474</td>\n",
       "      <td>...</td>\n",
       "      <td>36.568242</td>\n",
       "      <td>38.452451</td>\n",
       "      <td>37.429497</td>\n",
       "      <td>40.775615</td>\n",
       "      <td>35.785821</td>\n",
       "      <td>39.842322</td>\n",
       "      <td>32.569271</td>\n",
       "      <td>36.734122</td>\n",
       "      <td>35.965513</td>\n",
       "      <td>38.427939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>female_rate</th>\n",
       "      <th>248</th>\n",
       "      <td>0.509730</td>\n",
       "      <td>0.505057</td>\n",
       "      <td>0.514633</td>\n",
       "      <td>0.505058</td>\n",
       "      <td>0.513445</td>\n",
       "      <td>0.504086</td>\n",
       "      <td>0.498414</td>\n",
       "      <td>0.503256</td>\n",
       "      <td>0.496419</td>\n",
       "      <td>0.509407</td>\n",
       "      <td>...</td>\n",
       "      <td>0.503870</td>\n",
       "      <td>0.506135</td>\n",
       "      <td>0.491769</td>\n",
       "      <td>0.512958</td>\n",
       "      <td>0.497352</td>\n",
       "      <td>0.512202</td>\n",
       "      <td>0.474219</td>\n",
       "      <td>0.495299</td>\n",
       "      <td>0.520603</td>\n",
       "      <td>0.472677</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14 rows × 278 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "areaidx                                1            2           3    \\\n",
       "                       dateidx                                        \n",
       "newCasesRate           248        0.319707     0.778568    0.582971   \n",
       "                       249        0.426276     0.778568    0.510100   \n",
       "                       250        1.172258     0.637010    0.437229   \n",
       "                       251        1.065689     1.274021    0.291486   \n",
       "                       252        0.745982     0.566231    0.510100   \n",
       "                       253        1.278827     0.353895    0.145743   \n",
       "                       254        0.532845     0.353895    0.145743   \n",
       "People per sq. km      248      999.000000  2616.000000  560.000000   \n",
       "old_dependency_ratio   248        0.313702     0.258597    0.378299   \n",
       "young_dependency_ratio 248        0.288634     0.315276    0.280907   \n",
       "Average Score          248       35.000000    40.500000   29.800000   \n",
       "male_avg_age           248       40.157418    37.015974   42.037639   \n",
       "female_avg_age         248       42.190274    39.302367   44.115800   \n",
       "female_rate            248        0.509730     0.505057    0.514633   \n",
       "\n",
       "areaidx                                4           5            7    \\\n",
       "                       dateidx                                        \n",
       "newCasesRate           248        0.253268    0.465541     1.146148   \n",
       "                       249        0.253268    0.093108     1.480441   \n",
       "                       250        0.557191    0.558649     1.910247   \n",
       "                       251        0.557191    0.372433     1.910247   \n",
       "                       252        0.810459    0.651757     1.766979   \n",
       "                       253        0.557191    0.372433     1.480441   \n",
       "                       254        0.354576    0.093108     2.005759   \n",
       "People per sq. km      248      963.000000  541.000000  1163.000000   \n",
       "old_dependency_ratio   248        0.299317    0.333318     0.302608   \n",
       "young_dependency_ratio 248        0.299981    0.284279     0.279466   \n",
       "Average Score          248       25.800000   25.700000    18.900000   \n",
       "male_avg_age           248       39.646713   41.155788    40.452154   \n",
       "female_avg_age         248       41.726862   42.778638    42.309614   \n",
       "female_rate            248        0.505058    0.513445     0.504086   \n",
       "\n",
       "areaidx                                 8            9            10   \\\n",
       "                       dateidx                                          \n",
       "newCasesRate           248         0.933147     0.289057     0.038591   \n",
       "                       249         1.199760     0.578114     0.270139   \n",
       "                       250         2.132907     0.289057     0.540278   \n",
       "                       251         3.399320     0.289057     0.115774   \n",
       "                       252         2.199560     0.361321     0.115774   \n",
       "                       253         1.732987     0.361321     0.038591   \n",
       "                       254         1.732987     1.373021     0.077183   \n",
       "People per sq. km      248      1092.000000  3999.000000  3629.000000   \n",
       "old_dependency_ratio   248         0.229883     0.332554     0.231794   \n",
       "young_dependency_ratio 248         0.338812     0.285957     0.291240   \n",
       "Average Score          248        36.000000    45.000000    40.600000   \n",
       "male_avg_age           248        36.626540    41.063122    37.117334   \n",
       "female_avg_age         248        37.817297    42.667796    38.678151   \n",
       "female_rate            248         0.498414     0.503256     0.496419   \n",
       "\n",
       "areaidx                                11   ...          295          296  \\\n",
       "                       dateidx              ...                             \n",
       "newCasesRate           248        0.116550  ...     0.098261     0.242186   \n",
       "                       249        0.000000  ...     0.131015     0.193749   \n",
       "                       250        0.407924  ...     0.327537     0.193749   \n",
       "                       251        0.262237  ...     0.458552     0.290623   \n",
       "                       252        0.611886  ...     0.294783     0.193749   \n",
       "                       253        0.378787  ...     0.196522     0.435935   \n",
       "                       254        0.145687  ...     0.294783     0.193749   \n",
       "People per sq. km      248      142.000000  ...  8702.000000  5490.000000   \n",
       "old_dependency_ratio   248        0.452757  ...     0.135354     0.192277   \n",
       "young_dependency_ratio 248        0.261740  ...     0.272450     0.298154   \n",
       "Average Score          248       15.600000  ...    26.700000    14.600000   \n",
       "male_avg_age           248       44.593287  ...    35.124795    36.585779   \n",
       "female_avg_age         248       46.771474  ...    36.568242    38.452451   \n",
       "female_rate            248        0.509407  ...     0.503870     0.506135   \n",
       "\n",
       "areaidx                                 298          299           300  \\\n",
       "                       dateidx                                           \n",
       "newCasesRate           248         0.294447     0.252346      0.374980   \n",
       "                       249         0.425312     0.403753      0.156242   \n",
       "                       250         0.752475     0.656098      0.499973   \n",
       "                       251         0.392596     0.201876      0.468725   \n",
       "                       252         0.490745     0.353284      0.406228   \n",
       "                       253         0.556177     0.302815      0.437477   \n",
       "                       254         1.014205     0.100938      0.343732   \n",
       "People per sq. km      248      5411.000000  3450.000000  11045.000000   \n",
       "old_dependency_ratio   248         0.194712     0.251617      0.118766   \n",
       "young_dependency_ratio 248         0.320712     0.307247      0.241648   \n",
       "Average Score          248        17.200000     9.400000     25.800000   \n",
       "male_avg_age           248        35.694074    39.043512     34.824781   \n",
       "female_avg_age         248        37.429497    40.775615     35.785821   \n",
       "female_rate            248         0.491769     0.512958      0.497352   \n",
       "\n",
       "areaidx                                 301           302          303  \\\n",
       "                       dateidx                                           \n",
       "newCasesRate           248         0.048145      0.421726     0.288871   \n",
       "                       249         0.144434      0.512096     0.036109   \n",
       "                       250         0.192579      0.662712     0.324980   \n",
       "                       251         0.240724      0.632589     0.252762   \n",
       "                       252         0.096289      0.602466     0.397198   \n",
       "                       253         0.192579      0.421726     0.505525   \n",
       "                       254         0.288868      0.421726     0.541634   \n",
       "People per sq. km      248      4706.000000  16427.000000  7137.000000   \n",
       "old_dependency_ratio   248         0.238610      0.088241     0.161879   \n",
       "young_dependency_ratio 248         0.314247      0.257283     0.307642   \n",
       "Average Score          248        14.000000     27.900000    25.200000   \n",
       "male_avg_age           248        38.056436     32.614445    35.037225   \n",
       "female_avg_age         248        39.842322     32.569271    36.734122   \n",
       "female_rate            248         0.512202      0.474219     0.495299   \n",
       "\n",
       "areaidx                                 304           305  \n",
       "                       dateidx                             \n",
       "newCasesRate           248         0.121310      0.000000  \n",
       "                       249         0.242619      0.148232  \n",
       "                       250         0.636875      0.481753  \n",
       "                       251         0.363929      0.259405  \n",
       "                       252         0.303274      0.222347  \n",
       "                       253         0.667202      0.370579  \n",
       "                       254         0.454911      0.148232  \n",
       "People per sq. km      248      9624.000000  12167.000000  \n",
       "old_dependency_ratio   248         0.133132      0.175793  \n",
       "young_dependency_ratio 248         0.238689      0.223543  \n",
       "Average Score          248        16.600000     20.300000  \n",
       "male_avg_age           248        34.995129     36.918846  \n",
       "female_avg_age         248        35.965513     38.427939  \n",
       "female_rate            248         0.520603      0.472677  \n",
       "\n",
       "[14 rows x 278 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from typing import List\n",
    "from operator import add\n",
    "from toolz import reduce, partial\n",
    "from scipy.optimize import fmin_slsqp\n",
    "\n",
    "def loss_w(W, X, y) -> float:\n",
    "    return np.sqrt(np.mean((y - X.dot(W))**2))\n",
    "\n",
    "def get_w(X, y):\n",
    "    \n",
    "    w_start = [1/X.shape[1]]*X.shape[1] # initialized weights\n",
    "\n",
    "    weights = fmin_slsqp(partial(loss_w, X = X, y = y),\n",
    "                         np.array(w_start),\n",
    "                         f_eqcons = lambda x: np.sum(x) - 1,\n",
    "                         bounds=[(0.0, 1.0)]*len(w_start),\n",
    "                         disp=False)\n",
    "    return weights\n",
    "\n",
    "\n",
    "# only the pre-period\n",
    "\n",
    "features = [y_col, 'People per sq. km',\n",
    "            'old_dependency_ratio', 'young_dependency_ratio', 'Average Score', 'male_avg_age', \n",
    "           'female_avg_age', 'female_rate']\n",
    "\n",
    "inverted = (data.query(f\"~post_treatment\") # filter pre-intervention period\n",
    "            .pivot(index = 'areaidx' , columns= 'dateidx')[features] # make one column per year and one row per state\n",
    "            .T.drop_duplicates()) # flip the table to have one column per state\n",
    "\n",
    "inverted#.head()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "1841a303",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T23:44:42.460271Z",
     "start_time": "2022-04-14T23:44:42.450908Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "                        dateidx\n",
       "newCasesRate            248           1.258807\n",
       "                        249           0.699337\n",
       "                        250           1.338731\n",
       "                        251           1.178882\n",
       "                        252           1.858238\n",
       "                        253           1.838257\n",
       "                        254           1.738352\n",
       "People per sq. km       248        4453.000000\n",
       "old_dependency_ratio    248           0.214742\n",
       "young_dependency_ratio  248           0.241551\n",
       "Average Score           248          42.400000\n",
       "male_avg_age            248          36.823739\n",
       "female_avg_age          248          38.438335\n",
       "female_rate             248           0.499682\n",
       "Name: 249, dtype: float64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = inverted[treat_areaidx]# .values\n",
    "y\n",
    "#print(y.shape)\n",
    "# inverted[areaidx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "435eed64",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T23:44:17.886838Z",
     "start_time": "2022-04-14T23:44:17.879494Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([  1,   2,   3,   4,   5,   7,   8,   9,  10,  11,\n",
       "            ...\n",
       "            295, 296, 298, 299, 300, 301, 302, 303, 304, 305],\n",
       "           dtype='int64', name='areaidx', length=277)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = inverted.drop(columns=treat_areaidx)#.values  # other states\n",
    "# inverted.drop(columns=areaidx)\n",
    "X.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b42809a",
   "metadata": {},
   "source": [
    "# Get Weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "b423cfc3",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T23:44:56.384992Z",
     "start_time": "2022-04-14T23:44:55.368055Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "277\n",
      "Sum: 0.9999999999999999\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0.    , 0.4482, 0.    , 0.    , 0.    , 0.    , 0.0909, 0.3365,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.1244, 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    , 0.    ,\n",
       "       0.    , 0.    , 0.    , 0.    , 0.    ])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weights = get_w(X.values, y.values)\n",
    "\n",
    "print(len(weights))\n",
    "weights = np.round(weights, 4)\n",
    "print(\"Sum:\", weights.sum())\n",
    "weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "2ca910d2",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T23:55:36.803264Z",
     "start_time": "2022-04-14T23:55:36.796222Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{2: 0.4482, 8: 0.0909, 9: 0.3365, 284: 0.1244}\n",
      "[2, 8, 9, 284]\n",
      "[0.4482, 0.0909, 0.3365, 0.1244]\n"
     ]
    }
   ],
   "source": [
    "areidx2weight = {X.columns[idx]: weights[idx] for idx, weight in enumerate(weights) if weight > 0}\n",
    "print(areidx2weight)\n",
    "matched_idx = [i for i in areidx2weight]\n",
    "print(matched_idx)\n",
    "neat_weights = [areidx2weight[i] for i in matched_idx]\n",
    "print(neat_weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "e0d7b9ca",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T23:55:53.030340Z",
     "start_time": "2022-04-14T23:55:52.962694Z"
    },
    "scrolled": false
   },
   "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>areaidx</th>\n",
       "      <th>2</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>284</th>\n",
       "      <th>synth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.778568</td>\n",
       "      <td>0.933147</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.308347</td>\n",
       "      <td>0.569403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.778568</td>\n",
       "      <td>1.199760</td>\n",
       "      <td>0.578114</td>\n",
       "      <td>0.068522</td>\n",
       "      <td>0.661072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.637010</td>\n",
       "      <td>2.132907</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.308347</td>\n",
       "      <td>0.615015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.274021</td>\n",
       "      <td>3.399320</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.685215</td>\n",
       "      <td>1.062523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.566231</td>\n",
       "      <td>2.199560</td>\n",
       "      <td>0.361321</td>\n",
       "      <td>0.479651</td>\n",
       "      <td>0.634978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.353895</td>\n",
       "      <td>1.732987</td>\n",
       "      <td>0.361321</td>\n",
       "      <td>0.513912</td>\n",
       "      <td>0.501659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.353895</td>\n",
       "      <td>1.732987</td>\n",
       "      <td>1.373021</td>\n",
       "      <td>0.342608</td>\n",
       "      <td>0.820786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.283116</td>\n",
       "      <td>1.133107</td>\n",
       "      <td>0.144529</td>\n",
       "      <td>0.274086</td>\n",
       "      <td>0.312622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.283116</td>\n",
       "      <td>1.066453</td>\n",
       "      <td>0.578114</td>\n",
       "      <td>0.137043</td>\n",
       "      <td>0.435417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.637010</td>\n",
       "      <td>1.932947</td>\n",
       "      <td>0.867171</td>\n",
       "      <td>0.479651</td>\n",
       "      <td>0.812685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.495452</td>\n",
       "      <td>1.333067</td>\n",
       "      <td>1.011700</td>\n",
       "      <td>0.376868</td>\n",
       "      <td>0.730557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.424674</td>\n",
       "      <td>3.799240</td>\n",
       "      <td>1.083964</td>\n",
       "      <td>0.171304</td>\n",
       "      <td>0.921754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.344800</td>\n",
       "      <td>3.199360</td>\n",
       "      <td>1.011700</td>\n",
       "      <td>0.376868</td>\n",
       "      <td>1.280880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.707789</td>\n",
       "      <td>2.866093</td>\n",
       "      <td>1.589814</td>\n",
       "      <td>0.274086</td>\n",
       "      <td>1.146828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.061684</td>\n",
       "      <td>2.999400</td>\n",
       "      <td>0.505850</td>\n",
       "      <td>0.239825</td>\n",
       "      <td>0.948545</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "areaidx         2         8         9       284     synth\n",
       "0        0.778568  0.933147  0.289057  0.308347  0.569403\n",
       "1        0.778568  1.199760  0.578114  0.068522  0.661072\n",
       "2        0.637010  2.132907  0.289057  0.308347  0.615015\n",
       "3        1.274021  3.399320  0.289057  0.685215  1.062523\n",
       "4        0.566231  2.199560  0.361321  0.479651  0.634978\n",
       "5        0.353895  1.732987  0.361321  0.513912  0.501659\n",
       "6        0.353895  1.732987  1.373021  0.342608  0.820786\n",
       "7        0.283116  1.133107  0.144529  0.274086  0.312622\n",
       "8        0.283116  1.066453  0.578114  0.137043  0.435417\n",
       "9        0.637010  1.932947  0.867171  0.479651  0.812685\n",
       "10       0.495452  1.333067  1.011700  0.376868  0.730557\n",
       "11       0.424674  3.799240  1.083964  0.171304  0.921754\n",
       "12       1.344800  3.199360  1.011700  0.376868  1.280880\n",
       "13       0.707789  2.866093  1.589814  0.274086  1.146828\n",
       "14       1.061684  2.999400  0.505850  0.239825  0.948545"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_y = data.query(\"~treat_unit\").pivot(index='date', columns=\"areaidx\")[y_col]#[[i for i in areidx2weight]]\n",
    "area_synth = df_y[matched_idx].values.dot(np.array([areidx2weight[i] for i in matched_idx]))\n",
    "df_y = data.pivot(index='date', columns=\"areaidx\")[y_col]#[[i for i in areidx2weight]]\n",
    "synth_df = df_y[ matched_idx].reset_index(drop = True)\n",
    "synth_df['synth'] = area_synth\n",
    "synth_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "3b7f1b07",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-14T23:56:08.611542Z",
     "start_time": "2022-04-14T23:56:08.579909Z"
    }
   },
   "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>date</th>\n",
       "      <th>post_treatment</th>\n",
       "      <th>newCasesRate</th>\n",
       "      <th>2</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>284</th>\n",
       "      <th>synth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-09-05</td>\n",
       "      <td>False</td>\n",
       "      <td>1.258807</td>\n",
       "      <td>0.778568</td>\n",
       "      <td>0.933147</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.308347</td>\n",
       "      <td>0.569403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-09-06</td>\n",
       "      <td>False</td>\n",
       "      <td>0.699337</td>\n",
       "      <td>0.778568</td>\n",
       "      <td>1.199760</td>\n",
       "      <td>0.578114</td>\n",
       "      <td>0.068522</td>\n",
       "      <td>0.661072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-09-07</td>\n",
       "      <td>False</td>\n",
       "      <td>1.338731</td>\n",
       "      <td>0.637010</td>\n",
       "      <td>2.132907</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.308347</td>\n",
       "      <td>0.615015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-09-08</td>\n",
       "      <td>False</td>\n",
       "      <td>1.178882</td>\n",
       "      <td>1.274021</td>\n",
       "      <td>3.399320</td>\n",
       "      <td>0.289057</td>\n",
       "      <td>0.685215</td>\n",
       "      <td>1.062523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>False</td>\n",
       "      <td>1.858238</td>\n",
       "      <td>0.566231</td>\n",
       "      <td>2.199560</td>\n",
       "      <td>0.361321</td>\n",
       "      <td>0.479651</td>\n",
       "      <td>0.634978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2020-09-10</td>\n",
       "      <td>False</td>\n",
       "      <td>1.838257</td>\n",
       "      <td>0.353895</td>\n",
       "      <td>1.732987</td>\n",
       "      <td>0.361321</td>\n",
       "      <td>0.513912</td>\n",
       "      <td>0.501659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2020-09-11</td>\n",
       "      <td>False</td>\n",
       "      <td>1.738352</td>\n",
       "      <td>0.353895</td>\n",
       "      <td>1.732987</td>\n",
       "      <td>1.373021</td>\n",
       "      <td>0.342608</td>\n",
       "      <td>0.820786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>True</td>\n",
       "      <td>1.558523</td>\n",
       "      <td>0.283116</td>\n",
       "      <td>1.133107</td>\n",
       "      <td>0.144529</td>\n",
       "      <td>0.274086</td>\n",
       "      <td>0.312622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2020-09-13</td>\n",
       "      <td>True</td>\n",
       "      <td>1.598485</td>\n",
       "      <td>0.283116</td>\n",
       "      <td>1.066453</td>\n",
       "      <td>0.578114</td>\n",
       "      <td>0.137043</td>\n",
       "      <td>0.435417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2020-09-14</td>\n",
       "      <td>True</td>\n",
       "      <td>1.658428</td>\n",
       "      <td>0.637010</td>\n",
       "      <td>1.932947</td>\n",
       "      <td>0.867171</td>\n",
       "      <td>0.479651</td>\n",
       "      <td>0.812685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2020-09-15</td>\n",
       "      <td>True</td>\n",
       "      <td>1.838257</td>\n",
       "      <td>0.495452</td>\n",
       "      <td>1.333067</td>\n",
       "      <td>1.011700</td>\n",
       "      <td>0.376868</td>\n",
       "      <td>0.730557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2020-09-16</td>\n",
       "      <td>True</td>\n",
       "      <td>2.437689</td>\n",
       "      <td>0.424674</td>\n",
       "      <td>3.799240</td>\n",
       "      <td>1.083964</td>\n",
       "      <td>0.171304</td>\n",
       "      <td>0.921754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2020-09-17</td>\n",
       "      <td>True</td>\n",
       "      <td>3.216950</td>\n",
       "      <td>1.344800</td>\n",
       "      <td>3.199360</td>\n",
       "      <td>1.011700</td>\n",
       "      <td>0.376868</td>\n",
       "      <td>1.280880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2020-09-18</td>\n",
       "      <td>True</td>\n",
       "      <td>3.356818</td>\n",
       "      <td>0.707789</td>\n",
       "      <td>2.866093</td>\n",
       "      <td>1.589814</td>\n",
       "      <td>0.274086</td>\n",
       "      <td>1.146828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2020-09-19</td>\n",
       "      <td>True</td>\n",
       "      <td>3.316856</td>\n",
       "      <td>1.061684</td>\n",
       "      <td>2.999400</td>\n",
       "      <td>0.505850</td>\n",
       "      <td>0.239825</td>\n",
       "      <td>0.948545</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         date  post_treatment  newCasesRate         2         8         9  \\\n",
       "0  2020-09-05           False      1.258807  0.778568  0.933147  0.289057   \n",
       "1  2020-09-06           False      0.699337  0.778568  1.199760  0.578114   \n",
       "2  2020-09-07           False      1.338731  0.637010  2.132907  0.289057   \n",
       "3  2020-09-08           False      1.178882  1.274021  3.399320  0.289057   \n",
       "4  2020-09-09           False      1.858238  0.566231  2.199560  0.361321   \n",
       "5  2020-09-10           False      1.838257  0.353895  1.732987  0.361321   \n",
       "6  2020-09-11           False      1.738352  0.353895  1.732987  1.373021   \n",
       "7  2020-09-12            True      1.558523  0.283116  1.133107  0.144529   \n",
       "8  2020-09-13            True      1.598485  0.283116  1.066453  0.578114   \n",
       "9  2020-09-14            True      1.658428  0.637010  1.932947  0.867171   \n",
       "10 2020-09-15            True      1.838257  0.495452  1.333067  1.011700   \n",
       "11 2020-09-16            True      2.437689  0.424674  3.799240  1.083964   \n",
       "12 2020-09-17            True      3.216950  1.344800  3.199360  1.011700   \n",
       "13 2020-09-18            True      3.356818  0.707789  2.866093  1.589814   \n",
       "14 2020-09-19            True      3.316856  1.061684  2.999400  0.505850   \n",
       "\n",
       "         284     synth  \n",
       "0   0.308347  0.569403  \n",
       "1   0.068522  0.661072  \n",
       "2   0.308347  0.615015  \n",
       "3   0.685215  1.062523  \n",
       "4   0.479651  0.634978  \n",
       "5   0.513912  0.501659  \n",
       "6   0.342608  0.820786  \n",
       "7   0.274086  0.312622  \n",
       "8   0.137043  0.435417  \n",
       "9   0.479651  0.812685  \n",
       "10  0.376868  0.730557  \n",
       "11  0.171304  0.921754  \n",
       "12  0.376868  1.280880  \n",
       "13  0.274086  1.146828  \n",
       "14  0.239825  0.948545  "
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = data.reset_index(drop = True)\n",
    "# df['synth_y'] = area_synth\n",
    "df = df[['date', y_col, 'post_treatment']]\n",
    "df = data.query(\"treat_unit\").reset_index(drop = True)\n",
    "# df['synth_y'] = area_synth\n",
    "df = pd.concat([df[['date', 'post_treatment', y_col]], synth_df], axis = 1)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "e3a10c3a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-15T00:00:58.884008Z",
     "start_time": "2022-04-15T00:00:58.731291Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x16b745790>"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmQAAAFlCAYAAABFpfSEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAABvcElEQVR4nO3deXxM1//H8dfJTiKxRSxBqDUktohdKUVRLVVbW0V9u6rSftvqvvpWl++vC/12VbpRSquW2qu1FrHGErEFIYggiUT28/vjjjSIJJiZm5n5PB+PPJLcuXPve66RfHLOuecorTVCCCGEEMI8bmYHEEIIIYRwdVKQCSGEEEKYTAoyIYQQQgiTSUEmhBBCCGEyKciEEEIIIUwmBZkQQgghhMk8zA5wvSpXrqxDQkLMjiGEEEIIUawtW7ac0VoHFrefwxVkISEhREVFmR1DCCGEEKJYSqkjJdlPuiyFEEIIIUwmBZkQQgghhMmkIBNCCCGEMJnDjSErTHZ2NvHx8WRkZJgdRYhSw8fHh+DgYDw9Pc2OIoQQohhOUZDFx8dTrlw5QkJCUEqZHUcI02mtSUpKIj4+njp16pgdRwghRDGcossyIyODSpUqSTEmhIVSikqVKkmrsRBCOAinKMgAKcaEuIL8nxBCCMfhNAWZ2fz8/K7a9vnnn/Pdd9+ZkOba/vzzT/r27Wt2DCGEEEIU4BRjyEqrRx991CrHycnJwcND/qmEEEIIZyUtZDb0+uuv88EHHxATE0NkZGT+9ri4OMLCwgDYsmULt956K61ataJnz54kJCQA0KVLF8aNG0dERAQff/wxI0aM4NFHHyUiIoIGDRqwcOFCwBg/N3LkSMLCwmjRogWrVq0qcrsQQgghSh+na3Z5Y8Fu9pxIseoxQ6v789qdTW74+Y0aNSIrK4vDhw9Tp04dZs2axeDBg8nOzubJJ5/kt99+IzAwkFmzZvHSSy/xzTffAJCVlZW/TNSIESOIi4tj06ZNHDx4kK5du3LgwAE+/fRTlFJER0cTExNDjx49iI2NveZ2IYQQQpQ+TleQlVaDBg1i1qxZTJgwgVmzZjFr1iz27dvHrl27uP322wHIzc2lWrVq+c8ZPHjwVcdwc3Ojfv361K1bl5iYGNauXcuTTz4JGIVf7dq1iY2NveZ2IYQQQpQ+TleQ3UxLli0NHjyYe++9lwEDBqCUon79+kRHR9OkSRM2bNhQ6HN8fX0v+/7Ku+bkLjohrCM6PhmAsOAAk5MIIVyVjCGzk1tuuQV3d3feeuut/Javhg0bkpiYmF+QZWdns3v37mse4+effyYvL4+DBw9y6NAhGjZsSKdOnfjxxx8BiI2N5ejRo0VuF0JcLSw4QIoxIYSpnK6FzCzp6ekEBwfnf//0009ftc/gwYN59tlnOXz4MABeXl7MmTOHsWPHkpycTE5ODuPGjaNJk8Jb+WrVqkVkZCQpKSl8/vnn+Pj48Pjjj/PYY48RFhaGh4cH06dPx9vb+5rbhRBCCFH6KK212RmuS0REhL400P2SvXv30rhxY5MS2ceIESPo27cvAwcONDuKcCCu8H/DGj5cboyvHH97A5OTCCGcjVJqi9Y6orj9pMtSCOHyPl65n49X7jc7hhDChUmXpYOYPn262RGEcFpPdatvdgQhhIuTgkwI4fKkq1IIYTbpshRCCCGEMJkUZEIIlxcdn5w/F5kQQphBuiyFEC7vzilrAYib1MfkJEIIVyUtZFYyceJEmjRpQnh4OM2bN2fjxo03dJx58+axZ8+e/O+7dOnCldN8FCUuLo4ZM2bkfx8VFcXYsWNL/Pzs7GwmTJhA/fr1admyJe3atWPx4sUlfn5BV76Wkrre1yzEzWpaw5+mNfzNjiGEcGHSQmYFGzZsYOHChWzduhVvb2/OnDlDVlbWDR1r3rx59O3bl9DQ0Bt6/qWCbNiwYQBEREQQEVHs9Cf5XnnlFRISEti1axfe3t6cOnWKv/7664ayFPVacnJy8PCQt58oHRY+2cnsCEIIFyctZFaQkJBA5cqV82fCr1y5MtWrV+ePP/7g7rvvzt9v+fLl9O/fHwA/Pz9eeuklmjVrRtu2bTl16hTr169n/vz5PPvsszRv3pyDBw8CxpJJkZGRNGjQgDVr1gDGQuTPPvssrVu3Jjw8nC+++AKACRMmsGbNGpo3b86HH37In3/+Sd++fQG4cOECI0eOJCwsjPDwcObOnXvZ60hPT+err75i8uTJ+a8lKCiIQYMGATBz5kzCwsJo2rQpzz//fP7zSvpaunTpwrhx44iIiODjjz9m5cqVtGjRgrCwMEaNGkVmZqa1/2mEEEIIh+B8TRSLJ8DJaOses2oY3DHpmg/36NGDN998kwYNGtC9e3cGDx7MrbfeSteuXXn88cdJTEwkMDCQadOmMWrUKADS0tJo27YtEydO5LnnnuOrr77i5Zdfpl+/flfNyJ+Tk8OmTZv4/fffeeONN1ixYgVTp04lICCAzZs3k5mZSYcOHejRoweTJk3igw8+YOHChQD8+eef+cd56623CAgIIDrauD7nzp277HUcOHCAWrVq4e9/ddfNiRMneP7559myZQsVKlSgR48ezJs3j7vvvvu6XktWVhZRUVFkZGRQv359Vq5cSYMGDRg+fDifffYZ48aNu+5/HiGEEMLR2ayFTCnlo5TapJTaoZTarZR6o5B9RiilEpVS2y0fo22Vx5b8/PzYsmULX375JYGBgQwePJjp06ejlOKBBx7ghx9+4Pz582zYsIE77rgDMNaxvNRy1apVK+Li4q55/AEDBly137Jly/juu+9o3rw5bdq0ISkpif37i55pfMWKFTzxxBP531eoUKHEr3Hz5s106dKFwMBAPDw8uO+++1i9evV1v5ZLC6vv27ePOnXq0KCBMf/Tgw8+mH88IewtcuIKIieuMDuGEMKF2bKFLBO4TWt9QSnlCaxVSi3WWv99xX6ztNZjrHbWIlqybMnd3Z0uXbrQpUsXwsLC+PbbbxkxYgQjR47kzjvvxMfHh3vvvTd/3JSnpydKqfzn5uTkXPPYl7oPC+6ntWby5Mn07Nnzsn0Ltohdr3r16nH06FFSUlIKbSW7lut5Lb6+vjecTwhbOZ0q3eVCCHPZrIVMGy5YvvW0fDjWSuYltG/fvstap7Zv307t2rUBqF69OtWrV+ftt99m5MiRxR6rXLlypKamFrtfz549+eyzz8jOzgYgNjaWtLS0Ip9/++238+mnn+Z/f2WXZdmyZXnooYd46qmn8m9KSExMzB/D9tdff3HmzBlyc3OZOXMmt9566w2/loYNGxIXF8eBAwcA+P7774s9nhC2svHFbmx8sZvZMYQQLsymg/qVUu5Kqe3AaWC51rqwuSDuUUrtVErNUUrVvMZxHlZKRSmlohITE20Z+YZcuHCBBx98kNDQUMLDw9mzZw+vv/56/uP33XcfNWvWpHHjxsUea8iQIbz//vu0aNEif1B/YUaPHk1oaCgtW7akadOmPPLII+Tk5BAeHo67uzvNmjXjww8/vOw5L7/8MufOnaNp06Y0a9aMVatWXXXct99+m8DAQEJDQ2natCl9+/bF39+fatWqMWnSJLp27UqzZs1o1aoVd9111w2/Fh8fH6ZNm8a9995LWFgYbm5uPProo8VeHyFsIcjfhyB/H7NjCCFcmNLa9o1WSqnywK/Ak1rrXQW2VwIuaK0zlVKPAIO11rcVdayIiAh95RxVe/fuLVGxY5YxY8bQokULHnroIbOjCBdT2v9vCCGEs1NKbdFaFzv/lF2mvdBanwdWAb2u2J6ktb40eONroJU98thTq1at2LlzJ/fff7/ZUYQQ1/DCLzt54ZedZscQQrgwmw3qV0oFAtla6/NKqTLA7cC7V+xTTWudYPm2H7DXVnnMsmXLFrMjCCGKMXPTMQDeGRBuchIhhKuy5V2W1YBvlVLuGC1xs7XWC5VSbwJRWuv5wFilVD8gBzgLjLBhHiGEKNR/+oeZHUEI4eJsVpBprXcCLQrZ/mqBr18AXrBVBiGEKIlhbWqZHUEI4eJk6SQhhBBCCJNJQSaEcHkr9pxixZ5TZscQQrgwKcis5NSpUwwbNoy6devSqlUr2rVrx6+//mq380+fPp0xY6y34EFp0aVLF66c5gSMedj27NlT4uP8+OOPhIeHExYWRvv27dmxY0f+Y0uWLKFhw4bUq1ePSZP+Wenhvvvuo2HDhjRt2pRRo0blT8KrtWbs2LHUq1eP8PBwtm7dWug5MzMzGTx4MPXq1aNNmzb5S0plZWXlL/LerFmza66ukJSURNeuXfHz87vs3zY9PZ0+ffrQqFEjmjRpwoQJE0p8HUThRn8Xxejvrn6fCSGEvUhBZgVaa+6++246d+7MoUOH2LJlCz/99BPx8fFmR3NaX3/9NaGhoSXev06dOvz1119ER0fzyiuv8PDDDwOQm5vLE088weLFi9mzZw8zZ87ML/Tuu+8+YmJiiI6O5uLFi3z99dcALF68mP3797N//36+/PJLHnvssULPOXXqVCpUqMCBAwcYP348zz//PABfffUVANHR0SxfvpxnnnmGvLy8q57v4+PDW2+9xQcffHDVY//+97+JiYlh27ZtrFu3jsWLF5f4WoirdWtUhW6NqpgdQwjhwqQgs4I//vgDLy+vy2aar127Nk8++SQAcXFxdOrUiZYtW9KyZUvWr18PGOtO3nrrrdx1113UrVuXCRMm8OOPPxIZGUlYWFj+7PaJiYncc889tG7dmtatW7Nu3bpCc5w4cYJevXpRv359nnvuufztjz32GBERETRp0oTXXnstf3tISAjPPfccYWFhREZG5i9jNGLECB599FEiIiJo0KABCxcuBIzi5dlnn6V169aEh4fzxRdf5L+OLl26MHDgQBo1asR9993HpQmHQ0JCeO2112jZsiVhYWHExMQAkJaWxqhRo4iMjKRFixb89ttvAFy8eJEhQ4bQuHFj+vfvz8WLFwt9rQVbzvz8/HjppZdo1qwZbdu25dSpq7ue2rdvn7+Yetu2bfOL5U2bNlGvXj3q1q2Ll5cXQ4YMyc/Su3dvlFIopYiMjMx/zm+//cbw4cNRStG2bVvOnz9PQkLCVef87bffePDBBwEYOHAgK1euRGvNnj17uO02Y/7jKlWqUL58+UJbAX19fenYsSM+PpfPIF+2bFm6du0KGAu7t2zZUor/mzR1RGumjmhtdgwhhAtzyoIsZMIiQiYsumzbQ9M3EzJh0WXjRGZsPErIhEWXTQh5KiWDkAmLiJy4osTn2717Ny1btrzm41WqVGH58uVs3bqVWbNmMXbs2PzHduzYweeff87evXv5/vvviY2NZdOmTYwePZrJkycD8NRTTzF+/Hg2b97M3LlzGT16dKHn2b59O7NmzSI6OppZs2Zx7Jgxt9LEiROJiopi586d/PXXX+zc+c/rDQgIIDo6mjFjxjBu3Lj87XFxcWzatIlFixbx6KOPkpGRwdSpUwkICGDz5s1s3ryZr776isOHDwOwbds2PvroI/bs2cOhQ4cuKxorV67M1q1beeyxx/JbeyZOnMhtt93Gpk2bWLVqFc8++yxpaWl89tlnlC1blr179/LGG2+UaB63tLQ02rZty44dO+jcuXN+C9S1TJ06lTvuuAOA48ePU7PmPyt2BQcHc/z48cv2z87O5vvvv6dXr14lfs6V+3l4eBAQEEBSUhLNmjVj/vz55OTkcPjwYbZs2ZL/b3W9zp8/z4IFC+jWTdZhFEKUbskXs8nMyTU7Rqlly3nIXNYTTzzB2rVr8fLyYvPmzWRnZzNmzBi2b9+Ou7s7sbGx+fu2bt2aatWqAXDLLbfQo0cPAMLCwvLXmlyxYsVl46VSUlK4cOECfn5+l523W7duBAQEABAaGsqRI0eoWbMms2fP5ssvvyQnJ4eEhAT27NlDeLgxAebQoUPzP48fPz7/WIMGDcLNzY369etTt25dYmJiWLZsGTt37mTOnDkAJCcns3//fry8vIiMjCQ4OBiA5s2bExcXR8eOHQEYMGAAYKxa8MsvvwCwbNky5s+fn1+gZWRkcPToUVavXp1fsIaHh+fnLIqXlxd9+/bNP8fy5cuvue+qVauYOnUqa9euLfa4lzz++ON07tyZTp06lfg5RRk1ahR79+4lIiKC2rVr0759e9zd3a/7ODk5OQwdOpSxY8dSt25dq2QTQghrOnH+Ir9HJ7BwZwLbj50HIKCMJ1XKeRNYzrvAZ5+rvvcv44FSytwXYEdOWZDFTepz1bbCuiOGtal11fxDQf4+hT6/KE2aNGHu3Ln533/66aecOXOGiAhj6aoPP/yQoKAgduzYQV5e3mVdUN7e3vlfu7m55X/v5uZGTk4OAHl5efz9999XdV1dqeCx3N3d81tgPvjgAzZv3kyFChUYMWIEGRkZ+fsVfLNf6+tL32utmTx5Mj179rzssT///LPQc1+Zq+B2rTVz586lYcOGRb6mkvD09MzPe+W5C9q5cyejR49m8eLFVKpUCYAaNWpc1joVHx9PjRo18r9/4403SExMzO+eLeo5L730EosWGS2z27dvz98vODiYnJwckpOTqVSpEkqpyxZ+b9++PQ0aNODXX3/ljTfeAIwxcpfeP9fy8MMPU79+/ctaNsWNudSifr3/94UQV0tIvsjv0SdZtPMEW4+eB6BJdX/Gd2+Am4LEC5mcTskk8UImW46e43RKJpk5V4+j9fJwI9Dv2oVbFX9jW2U/bzzdHb/DzykLMnu77bbbePHFF/nss8/yB3inp6fnP56cnExwcDBubm58++235OZeX5Ntjx49mDx5Ms8++yxg/LJv3rx5iZ6bkpKCr68vAQEBnDp1isWLF9OlS5f8x2fNmsWECROYNWsW7dq1y9/+888/8+CDD3L48GEOHTpEw4YN6dmzJ5999hm33XYbnp6exMbGXla8XI+ePXsyefJkJk+ejFKKbdu20aJFCzp37syMGTO47bbb2LVr12Xdqzfj6NGjDBgwgO+//54GDRrkb2/dujX79+/n8OHD1KhRg59++okZM2YARlG0dOlSVq5ciZvbP//Z+/Xrx5QpUxgyZAgbN24kICCAatWqMXHiRCZOnHjZft9++y3t2rVjzpw53HbbbSilSE9PR2uNr68vy5cvx8PDg9DQUEJDQ+nfv3+JXs/LL79McnJy/o0GQghhppPJGSzelcCinQlEHTkHQGg1f57t2ZA+YdUIqex7zedqrUnNzDGKtNRMTqdmkJiamf9xOjWTI0npbI47y7n07EKPUdHXK79oC7xmq5s3ft6lt9VNCjIrUEoxb948xo8fz3vvvUdgYCC+vr68+66xdOfjjz/OPffcw3fffUevXr3w9b32G7Mwn3zyCU888QTh4eHk5OTQuXNnPv/88xI9t1mzZrRo0YJGjRpRs2ZNOnTocNnj586dIzw8HG9vb2bOnJm/vVatWkRGRpKSksLnn3+Oj48Po0ePJi4ujpYtW6K1JjAwkHnz5l3Xa7nklVdeYdy4cYSHh5OXl0edOnVYuHAhjz32GCNHjqRx48Y0btyYVq2ss978m2++SVJSEo8//jhgjOmKiorCw8ODKVOm0LNnT3Jzcxk1ahRNmjQB4NFHH6V27dr5heqAAQN49dVX6d27N7///jv16tWjbNmyTJs2rdBzPvTQQzzwwAPUq1ePihUr8tNPPwFw+vRpevbsiZubGzVq1OD777+/Zu6QkBBSUlLIyspi3rx5LFu2DH9/fyZOnEijRo3yxy6OGTPmmmMLRfGkZUyI63cqJYPF0Qn8Hn2SzUfOojU0qlqOf/doQO+watQN9Cv+IBi/Q/19PPH38aRelaKfk5WTx5kLRpFWsHg7XaB4O5SYxunUDLJz9VXP9/F0u6xQu61RFe6NqFnImexPXbobzlFEREToK+9I27t3L40bNzYpkeMKCQkhKiqKypUrX7Z9xIgR9O3bl4EDB5qUTFiL/N8QQljT6dQMluw6ycKdCWyOM4qwhkHl6BNejd5h1YotqOxFa8359OwC3aMZBVrg/inmeodV45keNz90pihKqS1a66LHoCAtZEIIIYQoQmJqJkt2G2PCNh42irAGQX6M69aAPuFVqVelnNkRr6KUooKvFxV8vWgQVPryFUYKMhd2aeb4K02fPt2uOYQw20PTNwOF3/wjhCs6cyGTJbtOsmhnAhsPJ5Gn4ZZAX8beVp8+4dUcpshxJFKQCSFc3sqY02ZHEMJ0SRcyWbr7FIuiT7DhoFGE1Q30ZUzXevQJr06DIL9SOyDeGThNQaa1ljeKEAU42vhQM309vNjhHUI4pbNpWSzdfZLfoxNYfzCJ3DxNncq+PNG1Hr3DqtGoajn53WonTlGQ+fj4kJSUlD/HkxCuTmtNUlJSsXPXCUP30CCzIwhhN+fSsli2xxiYf6kIC6lUlkdvrUufsOo0riZFmBmcoiALDg4mPj6exMREs6MIUWr4+Pjkr54ghHBtyenZLN1jjAlbd+AMOXmaWhXL8nDnuvQJq0aT6v5ShJnMKQoyT09P6tSpY3YMIYSDmrHxKMBVK3cI4ciS07NZtsfojlx74AzZuZqaFcswupNRhDWtIUVYaeIUBZkQQtyMF3+NBqQgE87jz32nefi7LWTl5lGjfBlGdahDn/BqhNUIkCKslJKCTAjh8oZGlo6ZuoWwhpzcPN5cuIfgimX4v0HNaRYsRZgjkIJMCOHy3hkQbnYEIazm5y3xHEpM44sHWtG8Znmz44gScvzl0YUQQggBwMWsXD5aEUvLWuXpIXcPOxQpyIQQLu9USganUjLMjiHETZu+Po5TKZk836uRdFM6GOmyFEK4vDb/WQlA3KQ+JicR4sYlp2fz2Z8H6NowkDZ1K5kdR1wnKciEEC6vSjlvsyMIcdP+99cBUjNzeK5XI7OjiBsgBZkQwuVteqm72RGEuCknkzOYvi6Ou5vXoHE1f7PjiBsgY8iEEEIIB/fxyljytObp2xuYHUXcICnIhBBCCAd2MPECs6Piua9NbWpWLGt2HHGDpCATQri8vpPX0HfyGrNjCHFDPli6Dx8PN8bcVs/sKOImyBgyIYTL23U8xewIQtyQ7cfOs3jXScZ1r09lP7k5xZFJQSaEcHkLxnQ0O4IQ101rzbuLY6jk68XoTnXNjiNukhRkQgiXFxYcYHYEIa7b6v1n2HAoidfvDMXPW36dOzoZQyaEEEI4mLw8o3UsuEIZhrWpbXYcYQVSkAkhXN6Hy2P5cHms2TGEKLEFO0+wJyGFZ3o0wMtDfpU7A5v9KyqlfJRSm5RSO5RSu5VSbxSyj7dSapZS6oBSaqNSKsRWeYQQ4lo+Xrmfj1fuNzuGECWSlZPHf5fF0qhqOe5qVsPsOMJKbNnpnAncprW+oJTyBNYqpRZrrf8usM9DwDmtdT2l1BDgXWCwDTMJIcRVnupW3+wIQpTYT5uPcvRsOtNGtMbNTRYQdxY2K8i01hq4YPnW0/Khr9jtLuB1y9dzgClKKWV5rhBC2MV4md1cOIi0zBw+WbmfyDoV6dIw0Ow4wops2vGslHJXSm0HTgPLtdYbr9ilBnAMQGudAyQDskS9EEIIUYipaw9z5kIWE+5ohFLSOuZMbFqQaa1ztdbNgWAgUinV9EaOo5R6WCkVpZSKSkxMtGpGIYSIjk8mOj7Z7BhCFCnpQiZfrj5Ej9AgWtaqYHYcYWXFFmRKqVaFbOt7PSfRWp8HVgG9rnjoOFDTckwPIABIKuT5X2qtI7TWEYGB0kQrhLCuO6es5c4pa82OIUSRPl11kPSsHJ7r1dDsKMIGStJC9lXBli2l1FDgleKepJQKVEqVt3xdBrgdiLlit/nAg5avBwJ/yPgxIYS9Na3hT9Ma/mbHEOKa4s+l88PfRxjYKph6VcqZHUfYQEkG9Q8E5iilhgGdgOFAjxI8rxrwrVLKHaPwm621XqiUehOI0lrPB6YC3yulDgBngSE38iKEEOJmLHyyk9kRhCjSh8v3g4Jx3eUGFGdVbEGmtT5kmZJiHnAU6KG1vliC5+0EWhSy/dUCX2cA915PYCGEEMKV7DuZyi/b4vlXp7pUL1/G7DjCRq5ZkCmlorl8moqKgDuwUSmF1jrc1uGEEEIIV/f+0hj8vD14vMstZkcRNlRUC9l1DdwXQghHFTlxBQCbXupuchIhLrc57iwr9p7m2Z4NKV/Wy+w4woauWZBprY8oY5KTSIz5wsC4K3KTDLwXQjiT06mZZkcQ4ipaGwuIVynnzagOdcyOI2ysqC7LHsD/gP0YhRgY84nVU0o9rrVeZod8Qghhcxtf7GZ2BCGusnLvaaKOnGNi/6aU8XI3O46wsaK6LD8Gumut4wpuVErVAX4HGtswlxBC2E2Qv4/ZEYS4TG6e5r2lMdSp7MugiJpmxxF2UNQ8ZB5AfCHbj2OsSymEEEIIG/h123FiT13g3z0a4ulu00V1RClRVAvZN8BmpdRPWNabxJhVfwjG/GFCCOEUXvhlJwDvDJCbx4X5MrJz+XB5LOHBAfQOq2p2HGEn1yy7tdbvAMMABbSzfCjgPstjQgjhFGZuOsbMTceK31EIO/jh7yMcP3+R53vJAuKupMiJYbXWe4G9dsoihBCm+E//MLMjCAFASkY2n646QMd6lelQr7LZcYQdlWTppKsopRZrre+wdhghhDDDsDa1zI4gBABfrT7EufRsnu/VyOwows6Kmvai5bUeAprbJI0QQgjhok6nZvD1msP0Ca9GWHCA2XGEnRXVQrYZ+AujALtSeZukEUIIE6zYcwqA7qFBJicRrmzyygNk5+bx7x4NzY4iTFBUQbYXeERrvf/KB5RSMvpVCOE0Rn8XBUDcpD4mJxGuKu5MGjM3HWVw65rUqexrdhxhgqIKste59l2YT1o/ihBCmKNboypmRxAu7r/LY/F0d+OpbvXNjiJMUtRalnOKeGyeTdIIIYQJpo5obXYE4cJ2HU9mwY4TPNH1FqrIqhEuS6b/FUIIIUz03tJ9lC/rySO33mJ2FGEiKciEEEIIk6w/eIbVsYk80aUe/j6yKqErK7IgU0q5KaXa2yuMEEKYIWTCIkImLDI7hnAxWmveXbKP6gE+PNCuttlxhMmKLMi01nnAp3bKIoQQQriMJbtOsuPYecbd3gAfT3ez4wiTlWSm/pVKqXuAX7TW2taBhBDC3mS6C2FvObl5vL9sH/Wr+HFPy2Cz44hSoCRjyB4BfgaylFIpSqlUpVSKjXMJIYQQTuvnLfEcSkzj2Z4NcXeTBcRFCVrItNbl7BFECCGEcAUXs3L5aEUsrWpX4HZZHUJYFNtCpgz3K6VesXxfUykVaftoQghhHw9N38xD0zebHUO4iOnr4ziVksnzvRqhlLSOCUNJxpD9D8gDbgPeAi5gDPSXmRSFEE5hZcxpsyMIF5Gcns1nfx7gtkZViKxT0ew4ohQpSUHWRmvdUim1DUBrfU4p5WXjXEIIYTdfD48wO4JwEf/76wCpmTk810sWEBeXK0lBlq2Ucgc0gFIqEKPFTAghnEJ3Gccj7CAh+SLT18XRv3kNGlX1NzuOKGVKcpflJ8CvQJBSaiKwFviPTVMJIYQQTubjFfvJ05rxtzcwO4oohUpyl+WPSqktQDfLpru11nttG0sIIexnxsajAAxrU8vkJMJZHTh9gdlRxxjeLoSaFcuaHUeUQiXpsgQoC1zqtixjuzhCCGF/L/4aDUhBJmzng6X7KOPpzpjb6pkdRZRSxRZkSqlXgXuBuYACpimlftZav23rcEIIYQ9DI2uaHUE4sW1Hz7Fk90nGda9PZT9vs+OIUqokLWT3Ac201hkASqlJwHZACjIhhFN4Z0C42RGEkzIWEI+hkq8XozvVNTuOKMVKMqj/BOBT4Htv4Lht4gghhBDOY/X+M/x96CxP3lYPP++SjhISrqgk745kYLdSajnGGLLbgU1KqU8AtNZjbZhPCCFs7lRKBgBB/j7F7ClEyeXlad5dHEPNimUY1qa22XFEKVeSguxXy8clf9omihBCmKPNf1YCEDepj8lJhDNZsPMEexJS+Ghwc7w8StIhJVxZSaa9+NYeQYQQwixVyslAa2FdWTl5/HdZLI2r+dOvWXWz4wgHYLMObaVUTeA7IAijq/NLrfXHV+zTBfgNOGzZ9IvW+k1bZRJCiMJseqm72RGEk/lp81GOnk1n2sjWuLnJAuKieLYcYZgDPKO13qqUKgdsUUot11rvuWK/NVrrvjbMIYQQQthNWmYOn6zcT5s6FenSINDsOMJBXFentlLKTSlVogW4tNYJWuutlq9Tgb1AjeuPKIQQQjiOqWsPc+ZCFs/f0QilpHVMlEyxBZlSaoZSyl8p5QvsAvYopZ69npMopUKAFsDGQh5up5TaoZRarJRqco3nP6yUilJKRSUmJl7PqYUQolh9J6+h7+Q1ZscQTiDpQiZfrj5EzyZBtKxVwew4woGUpIUsVGudAtwNLAbqAA+U9ARKKT+MWf7HWY5T0Fagtta6GTAZmFfYMbTWX2qtI7TWEYGB0vwrhLCuXcdT2HX8yh9PQly/T1cdJD0rh2d7NjQ7inAwJRlD5qmU8sQoyKZorbOVUrokB7c8by7wo9b6lysfL1igaa1/V0r9TylVWWt9pmTxhRDi5i0Y09HsCMIJxJ9L54e/j3Bvq5rUq1LO7DjCwZSkIPsCiAN2AKuVUrWBYv+UVEbH+VRgr9b6/66xT1XglNZaK6UiMVrskkqYXQghrCIsOMDsCMIJ/N/yWJSCcbfXNzuKcEAlmYfsE+CTApuOKKW6luDYHTC6NqOVUtst214EalmO+zkwEHhMKZUDXASGaK1L1PomhBBClBYxJ1P4ddtxHu5Ul2oBZcyOIxxQsQWZUioI+A9QXWt9h1IqFGiH0fp1TVrrtUCRt5doracAU0oeVwghrO/D5bEAjL+9gclJhCPKy9O89ttu/Lw9eKzLLWbHEQ6qJIP6pwNLgUtTDccC42yURwgh7O7jlfv5eOV+s2MIBzV17WE2Hj7Lq31DKV/Wy+w4wkGVZAxZZa31bKXUCwBa6xylVK6NcwkhhN081U3G/IgbE3MyhfeX7qNnkyAGtgo2O45wYCUpyNKUUpUwlj9CKdUWSLZpKiGEsCPpqhQ3IjMnl3E/bce/jCf/6R8mk8CKm1KSguxpYD5wi1JqHRCIMRhfCCGEcFn/tzyWmJOpfDMigkp+skC9uDkluctyq1LqVqAhxiD9fVrrbJsnE0IIO4mONxr9ZfoLUVIbDyXx5epDDGtTi9saBZkdRziBkiyddC9QRmu9G2Ny2FlKqZa2DiaEEPZy55S13DllrdkxhINIzcjm6dk7qF2xLC/1bmx2HOEkSnKX5Sta61SlVEegG8Z0F5/ZNpYQQthP0xr+NK3hb3YM4SDeWLCHhOSL/N/g5vh6l2TkjxDFK8k76dIdlX2Ar7TWi5RSb9swkxBC2NXCJzuZHUE4iCW7EpizJZ6xt9WTxcOFVZWkhey4UuoLYDDwu1LKu4TPE0IIIZzG6dQMXvglmrAaATwpU6UIKytJYTUIY2LYnlrr80BF4FlbhhJCCCFKE601z8/ZSXpWLh8Obo6nu7RLCOsq9h2ltU7XWv8CJCulagGeQIzNkwkhhJ1ETlxB5MQVZscQpdiMTUdZtS+RF3s3pl4VP7PjCCdUkrUs+wH/xVg66TTG4uAxQBPbRhNCCPs4nZppdgRRih0+k8bbC/fSqX5lHmhb2+w4wkmVZFD/W0BbYIXWuoVSqitwv21jCSGE/Wx8sZvZEUQplZObx/hZ2/HycOP9gc1wc5PZ+IVtlKQTPFtrnQS4KaXctNargAgb5xJCCLsJ8vchyN/H7BiiFPrfnwfZfuw8E/s3pWqAvEeE7ZSkhey8UsoPWA38qJQ6DaTZNpYQQghhrh3HzvPxyv3c3bw6fcOrmx1HOLmStJDdBaQD44ElwEHgTluGEkIIe3rhl5288MtOs2OIUuRiVi7jZ2+nSjlv3rirqdlxhAu4ZkGmlKqnlOqgtU7TWudprXO01t8CW4HydksohBA2NnPTMWZuOmZ2DFGKTFq8l0OJafz33mYElPE0O45wAUV1WX4EvFDI9mTLY9JKJoRwCv/pH2Z2BFGK/BWbyLcbjvBQxzq0r1fZ7DjCRRRVkAVpraOv3Ki1jlZKhdgukhBC2NewNrXMjiBKiXNpWTz78w7qV/Hj2Z4NzY4jXEhRBVn5Ih4rY+UcQgghhKm01rw8bxfn0rP4ZkRrfDzdzY4kXEhRg/qjlFL/unKjUmo0sMV2kYQQwr5W7DnFij2nzI4hTDZv+3EWRScw/vYGNK0RYHYc4WKKaiEbB/yqlLqPfwqwCMAL6G/jXEIIYTejv4sCIG5SH5OTCLMcP3+RV+ftJqJ2BR7pfIvZcYQLumZBprU+BbS3zMx/6Z7fRVrrP+ySTAgh7KRboypmRxAmysvTPDN7O3la83+DmuMus/ELExQ7MaxlZv5VdsgihBCmmDqitdkRhIm+WXeYvw+d5b17wqlVqazZcYSLKsnEsEIIIYRT2ncylfeW7OP20CDujQg2O45wYVKQCSGEcEmZObmMm7Ud/zIevDMgDKWkq1KYp0QFmVKqglLK39ZhhBDCDCETFhEyYZHZMYSdfbh8P3sTUpg0IJzKft5mxxEurqilk6orpb5TSiUDZ4BdSqmjSqnXlVKyjoQQQgiHtenwWb5YfZChkTXpHhpkdhwhimwh+wH4RmsdANwLzAUaY9wI8KkdsgkhhF3ETeojU164kNSMbJ6evZ2aFcrycp9Qs+MIARRdkFXSWv8JoLX+BehsWWj8ZaCzPcIJIYQQ1vbmgj2cOH+RDwc3w9e72MkGhLCLogqyRKXU/UqpGkqpJ4E4AGWMepSbAYQQQjicJbtO8vOWeB7vUo9WtSuaHUeIfEUVVqOAfsBSoA0wxrK9IvCCjXMJIYTdPDR9Mw9N32x2DGFjp1MzePHXaJrW8Gdst/pmxxHiMkXN1H8UGFTI9iSM8WRCCOEUVsacNjuCsDGtNRPmRpOWmcNHg5vj5SEdPaJ0uaHOc6XUq1rrN60dRgghzPD18AizIwgbm7npGH/EnOb1O0OpV6Wc2XGEuMqN/okwurgdlFI1lVKrlFJ7lFK7lVJPFbKPUkp9opQ6oJTaqZRqeYN5hBDihnUPDZKpD5xY3Jk03lq4h071KzO8XYjZcYQo1DVbyJRSKdd6CChTgmPnAM9orbcqpcoBW5RSy7XWewrscwdQ3/LRBvjM8lkIIYS4aTm5eYyfvR0vDzfeH9gMN1k4XJRSRbWQnQfqa639r/goByQUd2CtdYLWeqvl61RgL1Djit3uAr7Thr+B8kqpajf0SoQQ4gbN2HiUGRuPmh1D2MBnfx5k29HzvH13U6oG+JgdR4hrKmoM2XdAbeBUIY/NuJ6TKKVCgBbAxiseqgEcK/B9vGVbsQWfEEJYy4u/RgMwrE0tk5MIa9oZf56PV+7nrubVubNZdbPjCFGkou6yfLmIx54v6QmUUn4Yd2WO01pfqxu0uGM8DDwMUKuW/MAUQljX0MiaZkcQVnYxK5fxs7YTWM6bN/s1NTuOEMW6rrsslVKva61fv479PTGKsR8ts/1f6ThQ8CdhsGXbZbTWXwJfAkREROjrySyEEMV5Z0C42RGElb27JIaDiWn8OLoNAWVl+WVR+l3vXZb9SrqjZUb/qcBerfX/XWO3+cBwy92WbYFkrbV0VwohhLhhq2MTmb4+jlEd6tChXmWz4whRItc7D9n13J7SAXgAiFZKbbdsexGoBaC1/hz4HegNHADSgZHXmUcIIW7aqZQMAIL8ZdC3ozufnsWzc3ZQv4ofz/VqaHYcIUrseguyViXdUWu9lmIKOK21Bp64zgxCCGFVbf6zEoC4SX1MTiJuhtaal+bt4mxaFlMfbI2Pp7vZkYQosSK7LJVSXZVSv1gmdt0NzFZKdbFLMiGEsJMq5bypUs7b7BjiJs3fcYJFOxMY170BTWsEmB1HiOtS1MSwfYApwJvAGxitXS2Bb5RSY7TWv9snohBC2Naml7qbHUHcpBPnL/LyvF20ql2BR2+9xew4Qly3orosnwXu1lrvKLBtu1IqCpiMMf5LCCGEMFVenubfP+8gL0/z4aDmuMts/MIBFdVlWfWKYgwArfVOQBZ9E0IIUSpMWx/H+oNJvHpnKLUqlTU7jhA3pKiCLO0GHxNCCIfSd/Ia+k5eY3YMcQNiT6Xy7pIYujcOYlCETPArHFdRXZa3KKXmF7JdAXVtlEcIIexu1/EbWkREmCwrJ49xP22nnLcHk+4Jw5j+UgjHVFRBdlcRj31g7SBCOItDiRd4bs5OMnJyKevlgZ+3B2W93PH18sDX2wNfb3fKehmffb0Kfv/PtrJe7vh6e+Dt4Sa/ZOxgwZiOZkcQN+DDFbHsSUjhq+ERVPaTu2SFYytqLcu/7BlECGegtebFX6PZdyqV1iEVScvM4XRqBumZuaRl5ZCemcuFrBx0CRcAc3dTlPVy/6eos3w2vi9QwHl74OvlTllvD/wuFXheHpT1Nvb19fagqr+PDHa+hrBgmSLB0WyOO8vnfx1kSOua3B4qw5qF4ytq2oto4Jq/NrTWsvibEFf4ddtx/j50lv/0D2NYm1qF7qO1JiM7j7SsHNIyc0jLzCU9K4e0rFzL9zmkZ+VyITPH2H7p8QJF3YnzGfnPSc80PhennI8HEbUr0LpORdrUqUhYjfJ4eVzv6mlCmOtQ4gW+XR/HnC3x1KxQlpf7hpodSQirKKrLsq/dUgjhBJLTs5m4aC8tapVnSOtrDy5WSlHGy50yXu5W62bJy9NczM61FHm5+UVdWmYOaVk5pFzMIfr4eTYdPsuqfYkAeHu40bxmedrUqUjrOhVpWasCvt7Xu3iHc/hweSwA429vYHISUZi8PM3q/cb6lH/uS8TTXXFneHXGdquPn4u+Z4XzKeqd/BwwQ2u9zl5hhHBk7y2N4Vx6Ft89FImbnbsG3dyUZQyaB5S71l5Gi92ZC5lExZ1l0+FzbI47y5RVB8j7w+gebVrdn9YhFYmsU5HWIRWp4Otlt9dgpo9X7gekICtt0jJzmLs1nunr4ziUmEZlP2/Gda/PsDa1qFJO1h0VzqWogiwW+EApVQ2YDczUWm+zTywhHMv2Y+eZsekoI9vXoUn10j0eqbKfN72aVqNX02oApGZks/XoeTYfPsumw2f57u8jfL32MAD1q/gRWeefAq16+TJmRreZp7rVNzuCKOBoUjrfbohjdtQxUjNyaBYcwEeDm9M7rJp0swunpXQxo4uVUrWBIZaPMsBMjOIs1vbxrhYREaGjoqLMOLUQhcrJzeOuT9dx5kImK5/p4vBdKBnZuUQfT2aTpUDbcuQcFzJzAAiuUIbIEKOLM7JORepW9pW7QIVVaK3ZcDCJb9bFsTLmFO5KcUdYNUZ2CKFFzfLyPhMOSym1RWsdUex+xRVkVxy0BfANEK61dr+JfDdMCjJR2kxbd5g3Fuzh02Et6RNezew4VpeTm0fMydT8Am1z3FmS0rIAqOznRUTtivmtaI2r+cudnOK6XMzK5ddtx5m+/jCxpy5Q0deLYZG1uL9tbaoGSLekcHwlLciK/VNeKeUB3IHRQtYN+BN4/SbzCeEUTqVk8N9lsXRuEEjvsKpmx7EJD3c3mtYIoGmNAEZ1rIPWmkNn0vK7ODfFnWXJ7pMA+Hl70Kp2hfwuzvDgAHw8Tfnb7bpExycDMv2FPR0/f5HvNsTx06ZjJF/MpnE1f94bGE6/ZtUd4j0jhLUVNe3F7cBQoA+wEfgJeFhrLcsmCWHx1sI9ZOXm8Wa/Ji7TpaKU4pZAP24J9GNIpHGjwInzF9kcdza/Fe39pfsA8PJwo3lweaNAq1ORVrUrlMou3TunrAUgblIfk5M4N601m+POMW3dYZZaivieTaoyskMdWodUcJn/Q0IUpqifjC9gjBd7Rmt9zk55hHAYa/YnsnBnAuO7NyCksq/ZcUxVvXwZ7mpeg7ua1wDgbFqW5U5Oo4vzs78OMmXVAdwUNKkeQOuQinRvXIV2t1QqFb+Em9bwNzuCU8vIzmXBjhNMXx/H7hMpBJTx5F+d6/JA29oEV5DFwIWAYsaQKaW8gPuAJpZNuzGmwsi0Q7ZCyRgyURpkZOfS66PVKKVYMq4T3h7SxVKUtMwcth49Z3Rzxp1l29HzZObk0SDIjxHt69C/RQ3KeMk1dDanUjL4fsMRZmw6ytm0LPn3Fi7ppgf1K6UaAwuAdcAWy+ZWQAegn9Z6j5WyXhcpyERp8NGKWD5asZ8fHmpDx/qVzY7jcC61mExbF8eeBKPFZEhkTWkxcRJbj55j2ro4FkcnkKs13RoFMbJDCO1LSYuoEPZkjYJsJTBJa738iu3dgZe01l2tkvQ6SUEmzHb4TBo9P1pNzyZVmTy0hdlxHNqlMUXT1x9myS5jTFGP0KqM7BBCZJ2K8svbgWTl5PF7dALT1sex49h5ynl7MKh1TYa3q03tSq7dpS9cmzXusqxxZTEGoLVeoZSafFPphHBQWmte/W0X3u5uvNKnsdlxHJ5SKn/KjOPnL/L9hiPM3HSUJbtPElrNnxEdQuxy113kxBUAbHqpu03P44wSUzOZsfEoP2w8QmJqJnUr+/LmXU24p2Wwyy7FJcSNKOp/i5tSyvvK8WJKKZ9inieE01q4M4E1+8/wRr8mVPGXOZKsqUb5Mky4oxFPdavPvO3Hmb4ujufm7GTS4hibz0t1OtW0YbEOKzo+mWnrD7NwRwJZuXl0aRjIiPYhdK4faPelw4RwBkV1Wb4MtAWe0FofsWwLAT4BorTWb9orZEHSZSnMkpqRTbf//kWQvw/znuggE6Da2KWZ26etj2PF3n9mbh/RPoSWtaw7c/uplAwAgqTILlJ2bh5Ld59k2ro4thw5h6+XOwNbBTO8fQi3BPqZHU+IUummuyy11m8rpcYAa5RSl0bZpgEfaK2ly1K4nP8uiyXxQiZfPxghxZgdKKVoX68y7etV5mhSOt9tiGNW1DEW7DhBs+AARnQIoU9YdausbSiFWOEuZuVy5GwacWfS2JOQys9Rx0hIzqBWxbK80jeUeyOC8ffxNDumEE6hREsnKaXKAWitU22eqBjSQibMsOt4Mv2mrOW+NrV56+6mZsdxWWmZOfyyNZ5p6+M4lJhGYDlv7mtTi/va1CawnLfZ8RxSRnYuR8+mc/iMUXjFJaURdyaduKQ0EpIzLtu3Q71KjGxfh66NqsgfJUKUkDXusnwaSNZaT71i+0NAOa31R9YIer2kIBP2lpunGfDZeo6fu8jKZ24loIy0CJgtL0+z5sAZpq07zJ/7EvFyd6NveDVGdAghPLj8dR/vhV92AvDOgHArJy0dMrJzOWYpuo4kpXM4yVJ8nUkjISWDgr8GKvp6EVKpLCGVfalTyZfa+Z/LSmuYEDfAGndZ3ocxhuxK3wNRwEc3Fk0IxzJz01F2HDvPR4ObSzFWSri5KW5tEMitDQI5lHiBb9fHMWdLPL9sO06r2hUY0T6EXk2r4ulesu7MmZuOAY5dkGXl5HH0bHp+K1d+8XUmjRPJFy8rusqX9SSkki9t6lYipJIvIZXLWj77yntcCJMUVZB5aK2zr9yotc5SMjmQcBGJqZm8tySG9rdU4q7m1c2OIwpRN9CPN+5qyjM9GzInKp5vN8Tx5MxtVPX34YF2tRkaWYuKvl5FHuM//cPslPbmZOXkcexcOkeS0jh85vLi68T5i+QVKLr8fTyoU9mXiJAKhFQKpk5lo+AKqVSW8mWLvh5CCPsrqssyGuiutT51xfYgYIXW2pSfYNJlKezp6VnbWbDzBIuf6ky9KnIXmSPIzdP8ue8009bFsfbAGbw83Li7eXVGtK9DaPXSvWZlXp4mKS2LxNRMTqVkWMZzpXE4ySi+jp+/SG6BqqucpeiqXcmXOpZuxktdjeXLesrEukKUAtbosnwfWKSUegbYatnWyrL9g5uPKETptuFgEr9sO86YrvWkGHMg7m6Kbo2D6NY4iP2nUpm+Po5fth5ndlQ8bepUZGSHOtweGmTXQekZ2bkkpmZyOjWTxNQMy+dMTqdkknghk9OpGSSmZnLmQtZlBReAn7cHIZXLEh4cQL9m1Y2Cy9LFWNHXS4ouIZxEcYuL3wFMAJoCGmNx8Ula68X2iXc1aSET9pCVk0fvT9aQmZPL8vG32nymeGFbyenZzIo6yrfrj3D8/EVqlC/Dg+1rMziiFgFlPVmxx+gI6B4aVOJjaq05n55tFFQpmSReyDA+WwqvS0XW6dRMUjNyrnq+m4JKft4E+nlTxf+fz1XK+RBYzpsgf29qVfSlsp8UXUI4spu+y7K0koJM2MOnqw7w/tJ9TBvRmq6NqpgdR1hJTm4eK/aeZtq6w2w8fJYynu70b1mDGRuPAhA3qQ9ZOXmcuVB4YXXp8xnL11m5eVedw8fTjSrlfKhSzpvAct5UKedNFX8fAv28CSxQeFXy9ZapI4RwAdboshTCJR07m87kP/ZzR9OqUow5GQ93N3o1rUqvplXZcyKF6esPM2dLPABlPN1p8eYyzqVfdS8TYEwHcanIuiXQ11JsXV54BZbzxs/bQ1q0hBDXTQoyIQrQWvPa/N24K8Wrd4aaHUfYUGh1f94b2IwJdzRm5qajRMcnU8nPK7/L0GjZMoqsyn7eJZ5CQwghboQUZEIUsHT3Kf6IOc3LfRpTLaCM2XGEHVT09eKJrvXMjiGEcHHF/smnlDqolPpRKfWoUqpJSQ+slPpGKXVaKbXrGo93UUolK6W2Wz5evZ7gQlhbWmYObyzYTaOq5RjRPsTsOEIIIVxISdrgQ4EvgErA+5YC7dcSPG860KuYfdZorZtbPt4swTGFsJmPV+4nITmDif2b4iHdUy4lZMIiQiYsMjuGEMKFleS3Ti6QbfmcB5y2fBRJa70aOHtT6YSwk5iTKUxde5ghrWvSqnZFs+MIIYRwMSUZQ5YCRAP/B3yltU6y4vnbKaV2ACeAf2utdxe2k1LqYeBhgFq1alnx9EIYs6O//OsuAsp48nyvRmbHESaIm9TH7AhCCBdXkhayocBq4HHgJ6XUG0qpblY491agtta6GTAZmHetHbXWX2qtI7TWEYGBgVY4tRD/mLMlnqgj55hwRyMqFLPmoRBCCGELxRZkWuvftNbPAo8AvwMjgIU3e2KtdYrW+oLl698BT6VU5Zs9rhDX41xaFu8s3kvrkAoMbBlsdhwhhBAuqiR3Wc5VSh0APgbKAsOBCjd7YqVUVWWZPVEpFWnJYs3uUCGKNWlxDKkZObx9dxhuMmu6y3po+mYemr7Z7BhCCBdWkjFk7wDbtNa513NgpdRMoAtQWSkVD7wGeAJorT8HBgKPKaVygIvAEO1o6zgJhxYVd5ZZUcd4pHNdGlYtZ3YcYaKVMcXepySEEDZVkoJsD/CCUqqW1vphpVR9oKHWushuS6310GIenwJMKXlUIawnOzePl+ftonqAD2O71Tc7jjDZ18OLXWZOCCFsqiQF2TRgC9De8v1x4GesMI5MCLNMXxdHzMlUvnygFb7esmCFq+seGmR2BCGEiyvJXZa3aK3fw5iLDK11OiCDbYTDOnH+Ih+uiKV74yr0aFLV7DhCCCFEiQqyLKVUGUADKKVuATJtmkoIG3pjwW7ytOa1O0u8EphwcjM2HmXGxqNmxxBCuLCS9NW8BiwBaiqlfgQ6YEx9IYTD+SPmFEt3n+K5Xg2pWbGs2XFEKfHir9EADGsjE08LIcxRbEGmtV6ulNoKtMXoqnxKa33G5smEsLKLWbm8+ttu6lfxY3THumbHEaXI0MiaZkcQQri4axZkSqkr/1SMtnwua7njUtr3hUOZsmo/8ecu8tPDbfHykMXDxT/eGRBudgQhhIsrqoVsEca4sYID+DUQCFQB3G2YSwirOnA6lS9XH2JAyxq0rVvJ7DhCCCHEZa5ZkGmtwwp+r5QKAZ4HugP/sW0sUZrN2RLPvpMpDGgZTONq/mbHKZbWmpfn7aKslwcv9m5sdhxRCp1KyQAgyN/H5CRCCFdV7Bgyy0SwLwFtgP8CY7XW2bYOJkqnvQkpTJi7k5w8zVdrDhNWI4BBrWvSr1l1Asp4mh2vUPO2H+fvQ2eZ2L8plf28zY4jSqE2/1kJQNykPiYnEUK4qmsOpFFKNbUsfzQXWAE01Vp/LcWY68rJzeP5uTsJKOPJymdu5bU7Q8nOzeOVebuInLiCp37axroDZ8jLKz0rYCWnZzNx0V6a1yzP0NZyB50oXJVy3lQpJ8W6EMI8RbWQ7QCOYYwliwQiLWuBA6C1HmvbaKK0mbr2MDvjk5kyrAW3BPpxS6AfI9qHsPtECrOjjjFv23F+236CGuXLcG9EMANbBRNcwdypJd5fFsPZtCymj4yUxcPFNW16qbvZEYQQLq6ogmyU3VKIUu9Q4gX+b3ksPUKD6BNWLX+7UoqmNQJoWiOAF3s3Zunuk/wcFc9HK/bz8cr9dKxXmXsjatIjNAgfT/veB7L92Hl+3HiUEe1DaFojwK7nFkIIIa6H0rr0dC+VREREhI6KijI7hkvJy9MM+fJv9p5MYcXTt5Zo4POxs+nM3RrPz1HxHD9/EX8fD+5uUYNBETXtUhzl5mnu+nQtiamZrHj6Vsr5lM7xbUIIYRMntsOOn6BWG6h3O3j7mZ3IZSmltmitI4rbT1ZVLoTWmoLds67ux41H2BR3lvfuCS/xXWg1K5ZlXPcGjL2tPusPJjE76hg/bT7GdxuOEFrNn0ERwdzVvAYVfL1skvn7DXHsOp7Cp8NaSjEmitV38hoAFj7ZyeQkQlhBzO8w9yHIvggbPwN3b6jXDRr3g4a9oEwFsxOKQkhBdoWDiRcYP2s7/723GfWDypkdx3Tx59KZtDiGTvUrc29E8HU/381N0bF+ZTrWr0xyejbzdxxnVtQxXl+wh//8HsPtTYIYFFGTjvUq426lMV6nUjL4YFksnRsE0jtMFg8Xxdt1PMXsCEJYx9+fw5IJUL0FDJkBZw/B3gXGx77fwc0DQjpBaD9o2AfKBZmdWFgU22WplKqktU6yU55i2brL8lRKBr0/XkMlPy9+e6IjZbxcd/5brTUPTttMVNxZlo7rbNW1H3efSObnqHjmbT/O+fRsqgf4MLBVMANb1aRWpZs7z5gZW1m25xTLxnUmpLKvlRILZxYdnwxAWLCMNRQOKi8XlrwAm76ARn1hwFfgVeBnqdZwYqtRmO2ZD2cPAgpqtTVazhr3hfJyJ7otlLTLsiQF2X5gOzANWKxNHnRmjzFkq2MTeXDaJga1qsm7A113SZW5W+J55ucdvH5nKCM61LHJOTJzclmx5zSzo46xen8iWkO7upUY1DqYXk2qXXdBvGZ/Ig9M3cT47g14qnt9m2QWQohSJfMCzB0NsYuh3Ri4/U1wK+Jnp9Zweq+l5Ww+nNplbK/W3Gg5a9wPKsvPT2uxZkGmMGbnHwW0BmYD07XWsdYIer3sNaj//aUxfLrqIB8Nbs7dLWrY/HylzenUDG7/v9XUr+LH7Efa2WXKiBPnLzJ3Szw/b4nn6Nl0ynl70K95dQZF1CQ8OKDYcX0Z2bn0+mg1SimWjOuEt4frtm4KIVxESgLMGGQUVb3fh9ajr/8YSQchZqHRcnbc8vs1sJGl5exOqBoGMq76hlmtILvioF2BHwBfjHnKJmitN9xwyhtgr4IsJzePoV/9ze4TKSx4siO3BLrWHSqP/bCFlTGnWfxUJ7u/9rw8zcbDZ/k56hi/70ogIzuPhkHluDcimP4talDpGrPtf7Qilo9W7OeHh9rQsX5lu2YWju3D5cbfl+Nvb2ByEiGuw8ldRjGWkQwDp0GDHjd/zOTjRnG2dwEcWQc6DyqEGIVZ435QIwLcrjmnvCiENVvIKgH3Aw8Ap4CpwHygOfCz1to2fVnXYM9pLxKSL9L74zUE+fsw74kOdp9HyyyLoxN47MetPNerIY93qWdqlpSMbBbsOMHsqHh2HDuPp7uiW6MgBreuSaf6lfFwN34wxJ1Jo8dHq+nZpCqTh7YwNbNwPCETFgGydJJwIAdWwOwR4F0Ohs2CajYYXpN2BmIWGcXZoT8hLxvKVTPGqDW+E2p3AHe5N7A41izIYoHvgWla6/grHntea/3uTSW9Tvaeh2xVzGlGTt/MfW1qMbF/WPFPcHDn07Po/n+rqRrgzbzHO+QXPKXBvpOp/Bx1jF+2HedsWhZB/t7c0zKYeyNq8upvu9h+9Dwrn7mVKrJAtLhO0kImHErUN7Do31Al1CjGAuwwrCYjGWKXGmPO9q+AnItQpiI06m20nNXtAh6y/FhhrDqGzOyB/AWZMTHsO7/v5YvVh5gyrAV9w6vb9dz29vTs7czffoL5YzoSWt3f7DiFysrJ44+YU8yOiufPfae5tHSmLW8+EEII0+XlwYrXYP0nUL8HDPzGaCGzt6x0o4Vu7wKIXQKZKeBVDhr0NFrO6nWXiWgLuOmCTCm1ALhmIaa17nfj8W6cGQVZdm4eg77YwP5TF1j4ZEennUph1b7TjJy2mTFd6/Hvng3NjlMip1IymLMlnsTUTF7pG2q1ucyEEKJUyb4IvzxstFC1Hg293i0d3YU5WXD4LyNXzCJITwIPH6Moa3ynUaS5+ES01ijIbi3qiVrrv24w200xa+mk+HPp9PlkLTUrlmHuY+2d7g6+1Ixsen64mrLeHiwa29HpXp8QRZF5yESpdiERZg6B41ugx9vQ7onSeddjbg4c3fDPRLSpJ4yJaOt0Nro1G/UBvypmp7Q7m9xlWRqYuZblst0nefj7LYxoH8Lr/ZqYksFWXpm3ix82HmHOo+1pVdu1/5oRrkcG9YtSK3Ef/HgvXDgN93xltDo5grw8y0S0843pNM4dBuVmFGadnoZqzcxOaDc3vZalUmq21nqQUiqaQroutdYuN2NqjyZVGdWhDt+sO0zbuhXp1bSa2ZGsYuOhJL7/+wijOtSRYky4pKY1Sud4SeHiDq+GWfeDuxeMWATBrcxOVHJubhAcYXx0fwNO7YadsyBqGuyZZyx43ukZqN3O7KSlRlFdltW01glKqdqFPa61PmLTZNdgZgsZGAPKB36+nsNn0vh9bCerLidkhozsXO74eA25eZol4zpR1qsUjEkQQghXt30GzH8SKtWDYbOhQqG/ih3PxfOw+Sv4+zNjvFmt9kZhVq9b6eyGtYKStpBdc04DrXWC5fORwj6sGdaReHm4MWVoS9AwZuY2snLyzI50Uz5cHsvhM2lMGhAmxZgQQphNa/hjIsx7DEI6wqilzlOMAZQpD52fhXHRxo0J54/Aj/fAF51h96/GmpwuqthJppRSbZVSm5VSF5RSWUqpXKVUij3ClVa1KpXlvYHh7Dh2nveWxJgd54btOHaer9YcYmhkTdrXk5nthRDCVDmZ8Mu/YPV70OJ+uG+OUcA4Iy9faPsojN0O/aZAdjr8PAI+jYRtPxh3b7qYksz6OQUYCuwHygCjgU9tGcoR3BFWjeHtavP12sOs2HPK7DjXLSsnj+fn7qRKOR9e6N3Y7DhCmCpy4goiJ64wO4ZwZeln4bu7Ifpn6PaqUaS4e5qdyvY8vKDlA/DEJrh3OniWgd+egE9awN+fG3OeuYgSTcOutT4AuGutc7XW04Beto3lGF7s3ZjQav488/MOjp+/aHac6/K/Pw8QczKVt+9uir+PC/ynF6IIp1MzOZ2aaXYM4aqSDsLX3Y1pLe6ZaoypctLxVNfk5g5N+sMja4yWwfI1Ycnz8FFTWP2BMfbMyZWkIEtXSnkB25VS7ymlxpfweU7Px9OdT+9rSU5uHmNnbiM71zHGk+07mcqnqw7Qr1l1uocGmR3n5h3ZAMtfhewMs5MIB7XxxW5sfLGb2TGEKzr6t1GMXTwHD86HsIFmJzKXUlD/dhi1BEYuhuot4Y+34KMwWPG6Mf2HkypJYfWAZb8xQBpQE7jHlqEcSZ3KvrxzTzhbjpzjv8tizY5TrJzcPJ6bs4NyPp68dmeo2XFuXnwU/HAPrPsYZg93yXEH4uYF+fsQJGugCnuLngPf9jNmsh+9Amq1NTtR6VK7Pdw/Bx5ZDbfcBms/Mgqz35+F80fNTmd1xRZkljsqywHeWus3tNZPW7owhUW/ZtUZGlmLz/86yKp9pbt6n7Yujh3xybzerwmV/Bx8IdhTe+DHgeAXaIy52L8U5oyE3GyzkwkhxLVpbXTDzX0IarQyirFKt5idqvSq1gwGfQtjNhstiFHfGGPMfn0MEkt/Q0hJXbMgU4bXlVJngH1ArFIqUSn1akkOrJT6Ril1Wim1q4jjf6KUOqCU2qmUanljL6F0eO3OUBpVLcczs3dwMrl0dp3FnUnjg2X76N44iDvDHXxS23Nx8H1/cPeGB+YZYy7ueB9iFsLc0cYSHkKU0Au/7OSFX3aaHUO4gtxsmD/G6IYLuxeGz4OyFc1O5Rgq14e7PjXuzGw92pgm49NImPUAnNhmdrqbVlQL2XigA9Baa11Ra10BaAN0sIwjK850ih78fwdQ3/LxMPBZiRKXUj6e7kwZ1pKM7FzG/rSNnFI2niwvT/P83J14ebgxsX9TlCMPGE09Cd/dBTkZ8MCvULGOsb3Nw9BjojEL9LxHXXo+G3F9Zm46xsxNx8yOIZzdxfPGEIttP8Ctz8OAr8DDwXsqzFC+JtzxLozfZfwxfugv+LILfD8A4tYZLZAOqKiC7AFgqNb68KUNWutDwP3A8OIOrLVeDZwtYpe7gO+04W+gvFLKoZtt6lXx4+27m7Lp8Fk+Xrnf7DiXmbHpKBsPn+XlPo0de6xM+lmjZexCItw/F4KuGAfXfgx0e824dXz+k8Z6akIU4z/9w/hP/zCzYwhndu4IfNMTjqyHuz+Dri+63p2U1uZbGbq9AuOjjZ/7J3fC9N7GdY5d6nCFWVFTs3tqrc9cuVFrnaiUssY8CTWAgn+Sxlu2JVjh2KYZ0DKYDQeTmLLqAG3qVKJjffMnXD1x/iKTFsfQsV5lBkXUNDvOjcu8ADMGQdIBuO9nY420wnR6GnKz4M93jHl8+n4kP/hEkYa1qWV2BOHMjm+BGUMgNxMe+AXqdDY7kXPxCTB+7rd9zGh9XPex8bsiqCl0HG9Mp+HmbnbKYhXVQlbU7Wp2vZVNKfWwUipKKRWVmJhoz1PfkDfuakK9QD/GzdrG6VRzx5NprXnx12hy8zTvDAhz3K7KnExjkd3jW2DgN1C3S9H73/q80ZS9ZTosfs7h/lISQjiJvQtgWh9jwtOHlksxZkueZSDyXzB2m9EKmZtl3DgxJQK2fGv8HinFiirImimlUgr5SAWs0bZ/HGMKjUuCLduuorX+UmsdobWOCAwMtMKpbauslwef3teSC5k5jPtpO7l55hUDv247zp/7EnmuV0PHXQg9N8cYqH9olTF7deM7i3+OUnDbK9D+Sdj0JSx9SYoycU0r9pxyyBU3RCmmNayfYgw4D2oCo1dCYEOzU7kGd09oPgwe3wiDvgdvf1gwFj5uDhv+B1lpZicsVFGLi7trrf0L+SintbZGl+V8YLjlbsu2QPKlBc2dQYOgcrzZrynrDyYx5Q9zZglJTM3kzYV7aFW7AsPbhZiS4aZpDQufgr3zoec70OK+kj9XKbj9LWjzKPz9Kax8Q4oyUajR30Ux+rsos2MIZ5GbA7//G5a9ZPwBOWKhMT2PsC83NwjtBw//Cff/AhXrwtIX4MOm8Nd7xmS8pUhRY8huilJqJtAFqKyUigdeAzwBtNafA78DvYEDQDow0lZZzHJvRDDrD57h45WxRNapSLtbKtn1/K/P3016Zi7v3hOGu5sDdlVqDcteNsYEdH4O2j1+/cdQCnpNMpqu135oTJPR9QXrZxUOrVujKmZHEM4iMxV+HgkHlkP7sdD9DaMwEOZRCup1Mz6OboS1/werJhpjzbpMMHpSSgGbFWRa66HFPK6BJ2x1/tJAKcXb/cPYGZ/MUz9t4/enOlHZTpOxLtl1kkXRCTzbsyH1qpSzyzmtbs1/YcMUiHzYuCPpRikFvf9rFGV/TQJ3D+j8rPVyCoc3dURrsyMIZ5B8HGYMhtN7oO+HEDHK7ETiSrXawLBZcDLa+CPds4zZifJJ2W5jft4eTBnWkvMXsxk/azt5dhhPlpyezSu/7SK0mj8Pd65r8/PZxOavLRMnDoJe7978XZJubnDnJxA+GP54G9Z9Yp2cQggBxvxXX3czJq0eNluKsdKuaphxg1jr0WYnyScFmR2EVvfntTtDWbP/DJ/9ddDm53tr0R7OpmXx3sBwPN0d8J84eg4s+jc0uAPu/p/1mvvd3OGu/xm3QC9/Bf7+3DrHFUK4prw82LsQpvYw5r9S7sai2PW7m51MOCCbdVmKyw2LrMX6g0n833JjPFnrENsslfFXbCJztsTzRNdbaFojwCbnsKnYpfDrI1C7A9w7zbhbxprcPYzZsXOzYcnzxvFbP2TdcwiHEzJhEQBxk/qYnEQ4hOwM2PmTcRdl0n4oXwvueA9a3A9evmanEw7KAZtPHJNSikkDwgiuUIaxM7dxLs36U7ldyMzhxV+iuSXQlydvq2/149tc3DqYPdyYzG/oTNv17bt7wsBp0KAXLHoatn5vm/MIIZzLxXPGouAfhcGCp8CrLNwzFZ7cBm0ekWJM3BRpIbOjcj6efDqsJQP+t55nft7B18MjcLPi3Y/vLYnhRPJF5jzaHh/P0j8r8WVObIeZQ4y/NO+fCz7+tj2fhxcM+g5mDjWWWHL3hGZDbHtOUWpJy5go0vlj8Pf/jMlFs9Pglm7QYSzUuVVWARFWIy1kdta0RgAv9WnMHzGn+XrtIasdd9Phs3y34Qgj2ofQqnYFqx3XLs7sNxbc9QkwFgv3tdNyUx7eMORHqNMJ5j0Gu+ba57xCCMdwMhrm/gs+bmZMMN24Lzy61lj+qG4XKcaEVUkLmQmGt6vNhoNJvLdkHxEhFWlZ6+YKqIzsXJ6fu5PgCmV4tqeDzQR9/hh8d7fxg+2BeRAQbN/ze5aBoT/BDwONH7xunsZEgkII16Q1HPrTmKPq0Crw8jMml277GJR34LWARaknLWQmUErx7sBwqgb48OSMbSSnZ9/U8T5asZ/DZ9KYNCCcsl4OVGNfSITv7zYmUrz/F6hcz5wcXr5w32yo0QrmjIR9i83JIUzz0PTNPDR9s9kxhJlyc4w7vL/oZPxcOr0Hur0G43dBr/9IMSZsTgoykwSU8WTKsJacTs3g33N2oG9wSZ/o+GS+WnOIwRE16VjfTl191pCRDD8MMCZSHDYLqoWbm8e7HNw/B6qGGzcW7F9hbh5hVytjTrMy5rTZMYQZMi8YU+B80sJYiDo7A/pNhnHR0OlpKONgQ0CEw5KCzETNa5bn+V6NWL7nFNPWxV3387Ny8nh2zg4q+3nxYp/G1g9oK1npMGOI8Rfo4O+hdjuzExl8AoyxIYGN4KdhcHCV2YmcU14urJ8MO2aZnSTf18Mj+Hp4hNkxhD1dOA0r34IPmxhT4ATUgCEz4YlN0HK4McZUCDtyoP4t5/RQxzr8fSiJdxbvJSKkAuHB5Uv83M//OkjMyVS+Gh5BQBkrz9dlK7nZ8PODcHQDDJwK9W83O9HlylQwxrJ9e6dxB+b9cyCko9mpnMeFRKMV4vBfgDKud4MeZqeie2iQ2RFKt/go+PVR8PCBas2genPjc1BTY+oHR3LmAGyYDNtnGsupNeoDHZ6CmpFmJxMuTt1oV5lZIiIidFRUlNkxrOp8eha9P16Dh7sbC8d2xN+n+OIq9lQqfT5ZQ6+m1Zg8tIUdUlpBXi788jDsmgN9P4KIUrye/IVEmN4HkuONVrNabc1O5PiO/m0supyeZIzJ2fKtsczM6BUQ6GA3o7iS/cuNbnzfylCpPiRsN/4NAZQbVG74T4FWrZmxJI13KVw/99gmY6B+zCJw94LmQ6Hdk+aNXRUuQym1RWtdbBO8FGSlxJYjZxn0xd/0bBLEp8Naooq4nTo3T3PPZ+s5ejad5eM7U8lOC5bfFK1h0TMQNRW6vw4dx5udqHipJ2Fab6NrY/hvENzK7ESOSWvY8CmseA0Cahrzv1ULN+6w/aorePvDv1aaOlZnxsajAAxrU8u0DKXS9pkwfwxUCTXmB/SrYvx7ppwwCrOEHcYcggk74MJJy5MUVKp3eZFWrZkxJMDe8vIgdgms/8Rolfcpb6xd2OYR47UIYQdSkDmgz/48yLtLYnjrriY80C7kmvt9veYQby/ay8dDmnNX8xr2C3gzVr4Ja/5rdA3c/qbZaUou+bixRl36OXhwvvFLRpRcRjLMexxiFkKjvsbapAV/MR/9G6b3NbqF75tjLG1lAlk6qRDrPjHWfK1zKwz+ofjJmlNPGoVZwSItJf6fxyvUKVCkWT6Xtc0ScuRkws5ZxljFM7HGHwLtnoAWD4C3n23OKcQ1lLQgkzFkpcgjneuy8XASby3cS8vaFWhS/eq/KOPOpPHBsn10b1yFfs2qm5DyBqz7xCjGWj4I3d8wO831CagBDy4wWsq+vxseXAhVm5qdyjEk7DTGC547Aj0mGr8Qr2z5rdUW+n5otMIsf9XoyjTB0EiZ0iBfXp5RiG2YAk36Q/8vSjbAvVxV46NBz3+2XUiEkwUKtONbYPev/zxevlaBVrQWxme/wBvPfvE8RH0DGz+HC6eM7tMBX0OTu62/Lq4QViYtZKVM0oVMen+yhrJeHix4siN+3v/UzFprhn71N7uPp7D86VupGuBjYtIS2vqdsTRR6N0w8Btwc7AlnS45e9goynIzYcQiqOJAd7WaYev38Pu/jW7IgdOKv5N28QTY+Bnc9amxQLMwR242/PaE0boU+Qj0mgRuVr4ZP/0snNz5T5GWsB3OFli1xL9GgSKtufHZv1rRx0yOh78/gy3TIesC1O1qtMbLbPqiFJAuSwe28VASQ7/6m77h1fl4SPP88WQzNh7lxV+jmTQgjCGRDjDWZfc8Y6LVul2N2fA9vMxOdHOSDsK0O4wxNCN/h8oOuIC7rWWlw+/PwvYfjK6ue6aWrMUjNwd+vAeOrDdaIWu1sX1WcbnMC8bg/YMrodur0PFp+xUzGclGi+qlLs+E7caSalh+P/kFXV6gVWtmrOpxarcxPmzXXOP/ZdN7oP2T5s9rKEQBUpA5uMkr9/Pf5bH5xVdC8kVu/7/VhAcH8OPoNkUO+i8VDv4BPw6CGi2N9Sm9fM1OZB2J+4yWMndPo6Ws0i1mJyo9kg4av9BP7YLOz0GXCdfXIpp+Fr7uZhQGD6+y6zJap1IyAAjyd4BWZ1tIOwM/3msUQnd+bMzDZbbMC8ZakpcKtIQdkBgDOs943Kc8ZJwHT19o9aBlaSMH+ENVuBwpyBxcbp7mwW82sTnuLL+N6cB7S/ax4WASS8d1plalUj7vz7FN8N1dULEujFjofDNdn9ptDET3LGu0lFWobXYi8+35DeY9YQzKH/DVjc8vl7gPvuoGlerCyCV2m+PKpQf1nzsC3/eHlONG93Kj3mYnurasdOP/X8J2o9uzQgi0Gmm7mwOEsIKSFmQyU38p5e6m+HBwc/zLeDLky7/5I+Y0/+7ZsPQXYyd3wY8DjcG9D/zqfMUYQFATGD4PslKNCWST44t9itPKzYYlLxotY4EN4JE1NzfZb2BDY8LghJ3GWCY7/cFYpZw3Vco5wPQx1nZyF0ztYcwrNvy30l2MgVGg12wNkf8yljfq9IwUY8JpSEFWigWW8+bjwc1JvphNy1rlGdE+xOxIRUs6aKxP6elrzHbvzPP8VGtmFJwXzxlFWUqC2YnsL/m4MXnu358aA8BHLrHOAswNehpz1e3+xbg71w42vdSdTS91t8u5So24tcaYSDd3GLVEJj8WwmTSZekAth09R0glXyr4luJB8Skn4JuexriPkYuhSiOzE9nH0Y1GEepf3RhT5sxFaEEH/4C5o435nvp9YgymtiatjVUdomfDkBnG8jbCevbMN/79KoQYK1HYcbyeEK5GuiydSItaFUp3MZZ+1hiDkn7WWPvRVYoxMO4GHDbb6Lb8tp8xONqZ5eXBn+/C9wPAtwr8a5X1izEw7u7r9wlUb2kUZqf2WP8crmrzVGN+uGrNjJYxKcaEKBWkhay0SjtjLOgbvwmObwU3D6MVxr+GMVnppa/9a5g783RmqjGA/+Quoxir09m8LGY69BfMGGSs9ffgfOcc15KWBL/8y5gWIXywMaGrre+eTTkBX3Y1JiZ9+E+bXde+k9cAsPDJTjY5fqmgNfz1Lvz5DtTvCfdOd7yFwYVwQDJTvyPJzTamCoiPgvjNxl2K5w4bjyl3YxC5cjPuLEpLvPr53gGWAq26pVi7omDzr178sic3IjsDfhpmTPA4+AfXLcYA6t4KQ36EmUON1sLhv0GZ8mansp5jm41WlbREY2H4ViPsM0eVf3Xjuk7rbdw48MCvNplxfdfxFKsfs1TJyzXWkt0yDZrfZ0xtITPXC1GqSAuZGVJP/lN4xUfBiW2Qc9F4zC8IgltDzUjjc7Xml/8Vm5MJqQnGgOqUE8at6ikFvk4+Dmmnrz6nV7kCLWvVwT+4QNFmKeS8/Uv+SzY3x/gFHbPQWFql2ZCbvixOYd8SmHW/0R3U423j39FRVycAo1Vl4xew7GXjfTLoW6jewv45dvwEvz4Crf8FfT6w+uGj45MBCAs2YQFsW8vOgF9Gw94FxmSv3V6V2euFsCOZh6y0yMk0buGP3/zPR/Ix4zF3L+MXd3BrCI6A4EhjPMfN/rDMyTKKtisLtuR4y7YTxjpvXPFv7+VXSMFW3ch06Wuf8sYv6d+egB0z4I73oM0jN5fX2exdAHP/ZRTZvlWMAemN74SQTo61WkFGirHs1Z550OAO6P+ZudOYLHvFmJW974cQMcq8HI7k4nmjFfvIOmMZpLaPmZ1ICJcjBZkZtDaKnvhN/3Q/JuyA3Czj8YCa/xRewa2N5T1KsmivLeRkwYWTl7esXVnApZ7kqqLN09f4pZwSD11ehC7PmxK/1MtIgf3LjBbE2GWQnQY+AUZh0/hOuOW20j1+59Ruo4vw7GGjRaX9WOuvaXi98nJhxmA4tAqGz4eQDubmKe1SEow5ARP3Qf/PIWyg2YmEcElSkNlDVroxrqtg9+OFk8ZjHmWMrp3gCEsLWOviF8gtbXKzjZa05Cu7ReON19P+Sen6KInsi3DoT6PlbN/vxtxlHmWgfndo3A/q9yhd4822z4SF441xhwO/gZCOZif6R0ayMZP/xbPGHZ5WWiXhw+WxAIy/vYFVjme6MweMsYwXzxrjO2/panYiIVyWFGTWprUx0P5Yga7HU7sgL8d4vEKdAmO/IiCoqQyaFVfLzTa6j/YugL0LjQLezdO4KaDxndCwT8kW47aF7AxY/Bxs/dboXr1nKpQLMidLUc4cgK9vM1qcRy21yl3GTrV0UvwWmHEvoIw7n80Y8yeEyCcF2c3KTDWmmyg49is9yXjMy89YNDu4taX7MQJ8K9s+k3AueXlwfAvsnW98nIsz7qat1c4ozhr1tc7M9yVx9hDMftBYH7Dj09D1JWNdytLqwEqjO65hbxj0/U13pzpNC9mBFTBruPHz6IFfodItZicSwuVJQXajTu4y7uY6vQd0nrGtcoN/Cq/g1lClsWPfOSdKH62NcVt7Fxgfp3cb26s1N4qzxv2MtSJtIWYR/PoYKKD/l9Cwl23OY20b/gdLX4BbJ0DXF8xOY74ds+C3x42fT/fNLZ2tm0K4ICnIbtSFRKMgC25tLGJbo5VzLpAtSrekg/8UZ8ct7/fKDS3F2Z3G3bk3O34vNxtWvmncuVituTGlRYWQm01uP1rDb2Ng+w9w77fQ5G6zE5ln/RRY9pLR1Txkhm3mHRRC3BApyIRwFiknjFasvfMhbh3oXAio9U9xdiNznaUkwJxRcHQ9RDwEvd4x747fm5GTCdP7GuM5Ry017ly+AQ47D1leHqx4zSiqQ++CAV855r+jEE5MCjIhnFFaEsQuNlrODv5hTKniWwUa9bbMdda5+LnODq82irGsNGPG9vBB9sluK6mn4Kuuxvi7f626oZsiHHJQf2620UK48ydjwtw73pWhFEKUQqVicXGlVC+l1D6l1AGl1IRCHh+hlEpUSm23fIy2ZR4hHJ5vJWhxPwybBc8dskxL0QGi58AP98D79YzFuPcuMKZlKSgvD1Z/YKw9WqaCUbw4ejEGxlipITOM9V9nP2DMsXedmtbwp2kNB+rmy0ozluna+RN0fRl6vy/FmBAOzmYtZEopdyAWuB2IBzYDQ7XWewrsMwKI0FqPKelxpYVMiEJkZxSY62zR1XOdBbc2prTYvwya3gN3fmLuovS2sGuu0fLXcrjx+px1jry0JGNaixPbLOuKPmh2IiFEEUrD4uKRwAGt9SFLoJ+Au4A9RT5LCHH9PH2MuyMb9oLcj/+Z6yxmofEZjPnOen8ArUc7Z7HS9B44tQfWfABBYdDmYbMTWd/5o/D9AGP5tcE/GMtyCSGcgi0LshrAsQLfxwNtCtnvHqVUZ4zWtPFa62NX7qCUehh4GKBWrVo2iCqEE3H3MCaarXursdbo8S1w+C+o1835Jwnt+hKc3gtLJhjThNTtYnYi6zm12+iWzk6HB+ZB7XZmJxJCWJHJi9OxAAjRWocDy4FvC9tJa/2l1jpCax0RGGjSLOZCOCI3N2P6ls7/dv5iDIzXO+ALY+7A2Q8aE96WQOTEFUROXGHjcDfhyHr45g7j65FLpBgTwgnZsiA7DhScZjzYsi2f1jpJa51p+fZroJUN8wghXIF3ORg60+iWnTnUWOi9GKdTMzmdmlnsfqbYuxC+uxv8qsBDyyAo1OxEQggbsGVBthmor5Sqo5TyAoYA8wvuoJQquNp2P2CvDfMIIVxFxTow6Ds4sx9++Rfk5Ra5+8YXu7HxxW52CncdoqYZd45WDTPmWSsvQzaEcFY2K8i01jnAGGApRqE1W2u9Wyn1plKqn2W3sUqp3UqpHcBYYISt8gghXEydzsbcXLFL4I+3i9w1yN+HIH8fOwUrgbxc+Os9WDgO6nWHB+cbU54IIZyWTAwrhHBeWhtFzZbpcM9UCBtodqJru3jemOx3/zLjIz0Jmg2Dfp+Au6fZ6YQQN6g0THshhBDmUgrueB8SY+G3J6BiXajR8qrdXvhlJwDvDLixpZduiNZGl2rsEqMAO7oB8nKMSXvr3W6svhB6t3NOUSKEuIoUZEII5+bhBYO/hy+7wk/3wcOroFzVy3aZucmYbcfmBVlOJsStNQqw2CVwLs7YXqUJtB8LDXoak/jKrPtCuBwpyIQQzs+3MgydAVN7GEXZiEXGZLoW/+kfZrtzpyT80w15cBVkp4GHD9S51SjC6veA8jWLP44QwqlJQSaEcA1Vw6D/5zB7OCwcD3f/L787cFgbK969mJdnLGsUuwT2L4WEHcb2gJrQbIjRChbSCbzKWu+cQgiHJwWZEMJ1hN4FXV6AP9+BoCbQvsTL6BYtI8UYkB+7FA4sh7REUG4QHAndXjOKsCqhMh5MCHFNUpAJIVxL5+eMZYiWvwKBjaB+d1bsOQVA99Cgkh/nzIF/WsGOrDcG5PuUN6apaNDLWKqqbEXbvAYhhNORgkwI4Vrc3Iyuy6k9Yc4o+NdKRn8XC0DcpCIW687JMhZtj11qFGGXlmWqEgrtxhhFWHBrYy1RIYS4TvKTQwjherx8jUH+X3aFmUPoW/8DLrr7Xb1f6inLgPylxoD8rAvg7m1MOtv2caMrUmbPF0JYgRRkQgjXVL6WMR3Gt3cypfxkuO9nY0B+wvZ/WsFObDP29a8BYfcarWB1OsuAfCGE1UlBJoRwXbXbQ5//woKnYFpvoxsy7TSgoGYk3PaK0QoW1FQG5AshbEoKMiGEa2s1wpgxf/uPULerZUB+d1k7UghhV1KQCSFcXsiq9kB74u4tYlC/EELYkJvZAYQQQgghXJ20kAkhXF6R010IIYQdSAuZEEIIIYTJpCATQgghhDCZFGRCCCGEECaTgkwIIYQQwmRSkAkhhBBCmEwKMiGEEEIIk0lBJoQQQghhMinIhBBCCCFMJgWZEEIIIYTJpCATQgghhDCZFGRCCCGEECaTgkwIIYQQwmRSkAkhhBBCmEwKMiGEEEIIkymttdkZrotSKhE4YodTVQbO2OE8jkquT9Hk+hRPrlHR5PoUT65R0eT6FM8e16i21jqwuJ0criCzF6VUlNY6wuwcpZVcn6LJ9SmeXKOiyfUpnlyjosn1KV5pukbSZSmEEEIIYTIpyIQQQgghTCYF2bV9aXaAUk6uT9Hk+hRPrlHR5PoUT65R0eT6FK/UXCMZQyaEEEIIYTJpIRNCCCGEMJlDFGRKqZpKqVVKqT1Kqd1Kqacs2ysqpZYrpfZbPlewbL9PKbVTKRWtlFqvlGpW4Fi9lFL7lFIHlFITijjng5bj7ldKPVhg+2DLsXcrpd4t4vmtLOc/oJT6RCmlLNtfV0odV0ptt3z0lmt01TWaVeD6xCmltsv1uez6NFNKbbA8tkAp5X+z18eBr9FEpdQxpdSFK7Z3VkptVUrlKKUG3sx1KXBMZ7o+j1pybVdKrVVKhd7MtSlwXGe6RiOUUonqn59Fo2/m2liO6UzX58MC1yZWKXX+Ji5NweM60zWqrZRaaTnGn0qp4CJfvNa61H8A1YCWlq/LAbFAKPAeMMGyfQLwruXr9kAFy9d3ABstX7sDB4G6gBewAwgt5HwVgUOWzxUsX1cAKgFHgUDLft8C3a6ReRPQFlDAYuAOy/bXgX/LNbr2Nbpin/8Cr8r1uew9tBm41fL1KOAtF34PtbXkvnDF9hAgHPgOGCjX56rr41/g637AErlGV12jEcAUa1wXZ7w+V+zzJPCNXKOr3kM/Aw9avr4N+L7I127NN5u9PoDfgNuBfUC1Av+I+wrZtwJw3PJ1O2BpgcdeAF4o5DlDgS8KfP+FZVtrYGWB7Q8A/7vGGyqmsONho4LMma5RgW0KOAbUl+tz2XsomX/Gf9YE9rjie+iKYxX6ywKYjpUKMme8PgXOs1iuke0LMme6Plc8th64Xa7RVe+h3UBNy9cKSCnq+Q7RZVmQUioEaAFsBIK01gmWh04CQYU85SGM1gWAGhi/4C+Jt2y70rX2OwA0VEqFKKU8gLsxfiEW9vz4Is4zxtKE+c2lZldrcpJrBNAJOKW13l/I82+YE1yf3cBdlq/vvcbzb4qDXCPTOMP1UUo9oZQ6iNHyMPZ6n1+C44fg4NcIuMfys3qOUsqq70EnuT4opWoDdYA/buT5xRw7BMe+RjuAAZav+wPllFKVrrWzQxVkSik/YC4wTmudUvAxbZSg+or9u2L8Az1vjfNrrc8BjwGzgDVAHJB7nYf5DLgFaA4kYHTJWY2TXKNLhgIzrZHrEie5PqOAx5VSWzCa9LOske0SJ7lGNuMs10dr/anW+hZLrpetke0SJ7lGC4AQrXU4sByjy8oqnOT6XDIEmKO1tur/USe5Rv8GblVKbQNuBY4XdQyHKciUUp4Y/zg/aq1/sWw+pZSqZnm8GnC6wP7hwNfAXVrrJMvm41xe4QYDx5VSbQoMTux3rf0AtNYLtNZttNbtMJpQY5VS7gWe/6Zl3+BrPP+U1jpXa50HfAVE3uy1KfCaneIaWbJ5YPxlMetmrklBznJ9tNYxWuseWutWGAXrwZu9NgVesyNdI7tz0uvzE8Zf/1bhLNdIa52ktc60fPs10Op6r0VhnOX6FDAE6//h7BTXSGt9Qms9QGvdAnjJsu18UU8o9R8Yfa/fAR9dsf19Lh/k957l61oYzY3tr9jfA2PAXh3+GeTXpJDzVQQOY/RHV7B8XdHyWBX9T1/1dqDBNTJfOSC7t2V7tQL7jAd+kmt0+TWyPNYL+EveQ4W+hy49383ymka56jUqcCybjyFzputDgXGZwJ1AlFyjq65RwZ/V/YG/5fpcta0RRsuRssb7x9muEcbC5W6WrycCbxb5fGtdRFt+AB0xmid3Wi7KdqA3xl0QK4H9wIoCF/Fr4FyBfaMKHKs3xl0bB4GXijjnKMs/8gFgZIHtM4E9lo8hRTw/AthlOc+US29Y4Hsg2vJa5lPgP71co3/+U2P8In1U3kOFvoeespw/FpiElX4YOug1eg9jzEee5fPrlu2tLd+nAUnAbrk+l12fjzHGIm4HVlHILyq5RrxjuUY7LNeokVyff66P5bHXgUnWeO844zUCBlryxlpyehf12mWmfiGEEEIIkznMGDIhhBBCCGclBZkQQgghhMmkIBNCCCGEMJkUZEIIIYQQJpOCTAghhBDCZFKQCSGEEEKYTAoyIYQQQgiTSUEmhBBCCGGy/wcNHJDrJgwqZwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "\n",
    "time_idx = df['date']\n",
    "plt.plot(time_idx, df[y_col], label=treat_areaName)\n",
    "plt.plot(time_idx, df['synth'], label=\"Synthetic Control\")\n",
    "# for i in df[matched_idx]:\n",
    "#     plt.plot(time_idx, df[i], label='Control City {}'.format(i))\n",
    "    \n",
    "# plt.vlines(x = date, ymin=0, ymax = max(area_synth), linestyle=\":\", lw=2, label=str(date))\n",
    "plt.vlines(x = pd.to_datetime(treat_date), \n",
    "           ymin=0.9 * min(df[[y_col, 'synth']].min()), \n",
    "           ymax=1.1 * max(df[[y_col, 'synth']].max()), \n",
    "           linestyle=\":\", lw=2, label=\"Game happenned in {}\".format(treat_date))\n",
    "\n",
    "plt.ylabel(\"Daily New COVID-19 Cases per 10k\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "70f53f1e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-15T00:01:02.694097Z",
     "start_time": "2022-04-15T00:01:02.544301Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x16beee790>"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "\n",
    "time_idx = pd.to_datetime(data.query(\"treat_unit\")[\"date\"])\n",
    "\n",
    "diff = df[y_col] - area_synth\n",
    "plt.plot(time_idx, \n",
    "         diff,\n",
    "         label=\"Football Effect\")\n",
    "\n",
    "plt.vlines(x = pd.to_datetime(treat_date), \n",
    "           ymin=1.1 * min((diff)), ymax= 1.1* max(abs(diff)), \n",
    "           linestyle=\":\", lw=2, label=\"Game happenned in {}\".format(treat_date))\n",
    "\n",
    "plt.hlines(y=0, lw=2, xmin = min(time_idx), xmax = max(time_idx))\n",
    "plt.title(\"State - Synthetic Across Time\")\n",
    "plt.ylabel(\"Gap in per-capita cigarette sales (in packs)\")\n",
    "plt.legend()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "4abc083e",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-04-15T00:01:14.427703Z",
     "start_time": "2022-04-15T00:01:14.407797Z"
    }
   },
   "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>newCasesRate</th>\n",
       "      <th>synth</th>\n",
       "      <th>effect</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>post_treatment</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>1.415801</td>\n",
       "      <td>0.695062</td>\n",
       "      <td>0.720738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>2.372751</td>\n",
       "      <td>0.823661</td>\n",
       "      <td>1.549090</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                newCasesRate     synth    effect\n",
       "post_treatment                                  \n",
       "False               1.415801  0.695062  0.720738\n",
       "True                2.372751  0.823661  1.549090"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['effect'] = df[y_col] - df['synth']\n",
    "\n",
    "df.groupby('post_treatment')[[y_col, 'synth', 'effect']].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d65f6f44",
   "metadata": {},
   "source": [
    "# ---- Parking Lot "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "a7702569",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:30.459582Z",
     "start_time": "2022-03-18T14:31:30.453162Z"
    }
   },
   "outputs": [],
   "source": [
    "\n",
    "# import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "\n",
    "# required_gap = 7 # days\n",
    "\n",
    "\n",
    "# city_covid_df = ltla_covid_df[ltla_covid_df['areaName'] == city_name].reset_index(drop = True)\n",
    "# city_covid_df = city_covid_df.set_index('date')\n",
    "\n",
    "# fig, ax = plt.subplots(figsize=(15, 8))\n",
    "\n",
    "# game_date = pd.to_datetime(date)\n",
    "# gaps = pd.to_timedelta(required_gap, unit = 'D')\n",
    "# start = game_date - gaps\n",
    "# end = game_date + gaps\n",
    "\n",
    "\n",
    "# df = city_covid_df[start:end]\n",
    "# df[['cumCasesBySpecimenDate',\n",
    "#     # 'cumCasesBySpecimenDateRate', \n",
    "#     # 'newCasesBySpecimenDate'\n",
    "#    ]].plot(ax = ax)\n",
    "\n",
    "\n",
    "# xposition = [game_date]\n",
    "# for xc in xposition:\n",
    "#     ax.axvline(x=xc, color='grey', linestyle='--')\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "75540964",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-15T15:00:39.051186Z",
     "start_time": "2022-03-15T15:00:39.043185Z"
    }
   },
   "outputs": [],
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "73896d43",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:33.575858Z",
     "start_time": "2022-03-18T14:31:33.572139Z"
    }
   },
   "outputs": [],
   "source": [
    "# game_date = pd.to_datetime(date)\n",
    "# gaps = pd.to_timedelta(required_gap, unit = 'D')\n",
    "# start = game_date - gaps\n",
    "# end = game_date + gaps\n",
    "\n",
    "\n",
    "# def filter_time(date, start, end):\n",
    "#     return date > start and date < end\n",
    "\n",
    "\n",
    "# game_ltla_list = [i for i in ltla_game_df['team_LAD21NM'].unique() if i != ltla]\n",
    "# neighours = ltla_neighbours_df[ltla_neighbours_df['LAD21NM'] == ltla]['neighboors'].values[0]\n",
    "# print(game_ltla_list)\n",
    "# print(neighours)\n",
    "\n",
    "\n",
    "# df = ltla_covid_df[- ltla_covid_df['areaName'].isin(game_ltla_list + neighours)]\n",
    "# df = df[df['date'].apply(lambda x: filter_time(x, start, end))]\n",
    "# # df = df.rename(columns = {'cumCasesBySpecimenDate': 'cumcase'})\n",
    "# df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "51b77da1",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:35.859376Z",
     "start_time": "2022-03-18T14:31:35.856387Z"
    }
   },
   "outputs": [],
   "source": [
    "# df = df.reset_index(drop = True)\n",
    "# df['dateidx'] = df['date'].apply(lambda x: (x - game_date).days + required_gap)\n",
    "# df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "c7a973c6",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:38.931969Z",
     "start_time": "2022-03-18T14:31:38.928862Z"
    }
   },
   "outputs": [],
   "source": [
    "# final_df = pd.merge(df[['areaName', 'dateidx',  'date', 'cumcase', 'cumCasesBySpecimenDateRate', 'newCasesBySpecimenDate', ]], ltla_df, on = 'areaName')\n",
    "\n",
    "# final_df = final_df.sort_values(['areaidx', 'dateidx']).reset_index(drop = True)\n",
    "# final_df.to_csv('data/{}_{}.csv'.format(ltla, str(game_date.date())), index = False)\n",
    "# final_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86ba2aa7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "21240151",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "cdc85833",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-14T05:16:07.915428Z",
     "start_time": "2022-03-14T05:16:07.912119Z"
    }
   },
   "outputs": [],
   "source": [
    "# for areaName, ltla_df in df.groupby('areaName'):\n",
    "#     ltla_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "5f893c75",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-14T05:16:08.302374Z",
     "start_time": "2022-03-14T05:16:08.296585Z"
    }
   },
   "outputs": [],
   "source": [
    "# ltla_df['dateidx'] = list(range(len(ltla_df)))\n",
    "# ltla_df = ltla_df.rename(columns = {'cumCasesBySpecimenDate': 'cumcase'}).reset_index(drop = True)\n",
    "# ltla_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "d1dbae0a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-14T05:16:08.717419Z",
     "start_time": "2022-03-14T05:16:08.713691Z"
    }
   },
   "outputs": [],
   "source": [
    "# ltla_df.pivot_table(values = 'cumcase', index = 'areaName', columns = 'dateidx').reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "2419a143",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:54.556366Z",
     "start_time": "2022-03-18T14:31:54.553022Z"
    }
   },
   "outputs": [],
   "source": [
    "# df['dateidx'] = df['date'].apply(lambda x: (x - game_date).days )\n",
    "# df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "84bded52",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:57.029684Z",
     "start_time": "2022-03-18T14:31:57.026900Z"
    }
   },
   "outputs": [],
   "source": [
    "# ltla_covid_flatten_df = df.pivot_table(values = 'cumcase', index = 'areaName', columns = 'dateidx').reset_index()\n",
    "# ltla_covid_flatten_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "6e83764f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:31:59.160990Z",
     "start_time": "2022-03-18T14:31:59.157999Z"
    }
   },
   "outputs": [],
   "source": [
    "# final_df = pd.merge(ltla_covid_flatten_df, ltla_df, on = 'areaName')\n",
    "# final_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "9d638baa",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-03-18T14:32:01.373968Z",
     "start_time": "2022-03-18T14:32:01.370909Z"
    }
   },
   "outputs": [],
   "source": [
    "# final_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8b949226",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "165px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}