vaccine_efficiency_bulgaria.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0a06ce85",
   "metadata": {},
   "source": [
    "This is an unpretentions attempt to empirically estimate the COVID-19 vaccine efficiency in Bulgaria in preventing death. It is not a diligent statistical analysis, but rather an attempt to explore the variance in the daily reports.\n",
    "\n",
    "Unfortunately, there is not much that can be done with the available data. This notebook contains the following sections (hyperlinks are broken in GitHub; scroll down):\n",
    "\n",
    "1) The [data loading section](#Data-loading) contains code for fetching and aggregating data from multiple governmental sources.\n",
    "\n",
    "2) The [historical data section](#Efficiency-estimates-based-on-historical-data) contains an overview of how the overall estimated efficiency evolves with time.\n",
    "\n",
    "3) The [regional distribution section](#Efficiency-estimates-based-on-regional-distribution) contains an estimated regional efficiency distribution.\n",
    "\n",
    "We are working with ratios of ratios of heavily skewed distributions, which makes it particularly challenging to build a theoretical model. Furthermore, the data is incomplete in the official governmental sources, and I try to fill any missing bits - see the introductory notes in the individual sections for details."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41929533-5d90-44fa-b2e0-356e2cb4231d",
   "metadata": {},
   "source": [
    "# Data loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f8d2bf0f-14c6-47e0-a09f-138db775becd",
   "metadata": {},
   "outputs": [],
   "source": [
    "from bs4 import BeautifulSoup\n",
    "from typing import Callable, List\n",
    "import io  # Feed strings into pandas\n",
    "import itertools\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import plotnine as p9\n",
    "import pywt\n",
    "import requests\n",
    "import scipy  # Nonparametric confidence intervals\n",
    "import warnings\n",
    "\n",
    "# Loading statsmodels before pandas causes deprecation warnings\n",
    "from statsmodels.nonparametric.api import lowess\n",
    "from statsmodels.formula.api import wls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8f0136b0-27b6-431e-b326-da6740811247",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>7d</th>\n",
       "      <th>7d-100</th>\n",
       "      <th>14d</th>\n",
       "      <th>14d-100</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Благоевград</th>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>31</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Бургас</th>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>43</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Варна</th>\n",
       "      <td>27</td>\n",
       "      <td>5</td>\n",
       "      <td>58</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Велико Търново</th>\n",
       "      <td>17</td>\n",
       "      <td>7</td>\n",
       "      <td>37</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Видин</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                7d  7d-100  14d  14d-100\n",
       "Благоевград      8       2   31       10\n",
       "Бургас          15       3   43       10\n",
       "Варна           27       5   58       12\n",
       "Велико Търново  17       7   37       16\n",
       "Видин            4       4   10       12"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cumulative deceased by region in 7 and 14 day periods\n",
    "deceased_json = requests.get('https://coronavirus.bg/statistics/deceased-7-14.json').json()\n",
    "deceased_df = pd.DataFrame.from_records(deceased_json['data']).transpose().astype(int).sort_index()\n",
    "regions = deceased_df.index\n",
    "\n",
    "deceased_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6f89ef4-78bb-48b6-8700-31eea167a9be",
   "metadata": {},
   "source": [
    "### Data per region"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0e75f2ba-d558-4c54-9b7b-b9adff5294e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Общо поставени дози</th>\n",
       "      <th colspan=\"4\" halign=\"left\">Нови</th>\n",
       "      <th>Общ брой лица със завършен ваксинационен курс</th>\n",
       "      <th>Общ брой лица с поставена бустерна доза (реваксинация)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Общо поставени дози</th>\n",
       "      <th>Comirnaty*</th>\n",
       "      <th>Moderna**</th>\n",
       "      <th>AstraZeneca***</th>\n",
       "      <th>Janssen****</th>\n",
       "      <th>Общ брой лица със завършен ваксинационен курс</th>\n",
       "      <th>Общ брой лица с поставена бустерна доза (реваксинация)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>(Област, Област)</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>Благоевград</th>\n",
       "      <td>150333</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>-</td>\n",
       "      <td>3</td>\n",
       "      <td>72453</td>\n",
       "      <td>22941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Бургас</th>\n",
       "      <td>255413</td>\n",
       "      <td>25</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>5</td>\n",
       "      <td>120442</td>\n",
       "      <td>41057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Варна</th>\n",
       "      <td>303163</td>\n",
       "      <td>64</td>\n",
       "      <td>14</td>\n",
       "      <td>-</td>\n",
       "      <td>10</td>\n",
       "      <td>147092</td>\n",
       "      <td>47344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Велико Търново</th>\n",
       "      <td>114280</td>\n",
       "      <td>5</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>1</td>\n",
       "      <td>53585</td>\n",
       "      <td>20022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Видин</th>\n",
       "      <td>36635</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>-</td>\n",
       "      <td>17160</td>\n",
       "      <td>6437</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Общо поставени дози       Нови                           \\\n",
       "                 Общо поставени дози Comirnaty* Moderna** AstraZeneca***   \n",
       "(Област, Област)                                                           \n",
       "Благоевград                   150333         20         1              -   \n",
       "Бургас                        255413         25         -              -   \n",
       "Варна                         303163         64        14              -   \n",
       "Велико Търново                114280          5         -              -   \n",
       "Видин                          36635          -         -              -   \n",
       "\n",
       "                             Общ брой лица със завършен ваксинационен курс  \\\n",
       "                 Janssen**** Общ брой лица със завършен ваксинационен курс   \n",
       "(Област, Област)                                                             \n",
       "Благоевград                3                                         72453   \n",
       "Бургас                     5                                        120442   \n",
       "Варна                     10                                        147092   \n",
       "Велико Търново             1                                         53585   \n",
       "Видин                      -                                         17160   \n",
       "\n",
       "                 Общ брой лица с поставена бустерна доза (реваксинация)  \n",
       "                 Общ брой лица с поставена бустерна доза (реваксинация)  \n",
       "(Област, Област)                                                         \n",
       "Благоевград                                                   22941      \n",
       "Бургас                                                        41057      \n",
       "Варна                                                         47344      \n",
       "Велико Търново                                                20022      \n",
       "Видин                                                          6437      "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cumulative vaccinated by region\n",
    "vaccine_soup = BeautifulSoup(requests.get('https://coronavirus.bg/bg/statistika').text)\n",
    "vaccine_df = pd.read_html(str(vaccine_soup.find_all('table')[5]))[0] \\\n",
    "    .set_index(('Област', 'Област')) \\\n",
    "    .drop('Общо') \\\n",
    "    .sort_index()\n",
    "\n",
    "vaccine_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e5925828-674e-4dd0-8315-14cbd8296dcf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">Общо</th>\n",
       "      <th colspan=\"3\" halign=\"left\">В градовете</th>\n",
       "      <th colspan=\"3\" halign=\"left\">В селата</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>всичко</th>\n",
       "      <th>мъже</th>\n",
       "      <th>жени</th>\n",
       "      <th>всичко</th>\n",
       "      <th>мъже</th>\n",
       "      <th>жени</th>\n",
       "      <th>всичко</th>\n",
       "      <th>мъже</th>\n",
       "      <th>жени</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>(ОбластиОбщини, ОбластиОбщини)</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>Благоевград</th>\n",
       "      <td>301 138</td>\n",
       "      <td>146 369</td>\n",
       "      <td>154 769</td>\n",
       "      <td>181 513</td>\n",
       "      <td>86 646</td>\n",
       "      <td>94 867</td>\n",
       "      <td>119 625</td>\n",
       "      <td>59 723</td>\n",
       "      <td>59 902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Бургас</th>\n",
       "      <td>409 750</td>\n",
       "      <td>197 692</td>\n",
       "      <td>212 058</td>\n",
       "      <td>311 247</td>\n",
       "      <td>148 393</td>\n",
       "      <td>162 854</td>\n",
       "      <td>98 503</td>\n",
       "      <td>49 299</td>\n",
       "      <td>49 204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Варна</th>\n",
       "      <td>470 124</td>\n",
       "      <td>228 625</td>\n",
       "      <td>241 499</td>\n",
       "      <td>390 916</td>\n",
       "      <td>188 931</td>\n",
       "      <td>201 985</td>\n",
       "      <td>79 208</td>\n",
       "      <td>39 694</td>\n",
       "      <td>39 514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Велико Търново</th>\n",
       "      <td>229 718</td>\n",
       "      <td>110 803</td>\n",
       "      <td>118 915</td>\n",
       "      <td>160 186</td>\n",
       "      <td>76 540</td>\n",
       "      <td>83 646</td>\n",
       "      <td>69 532</td>\n",
       "      <td>34 263</td>\n",
       "      <td>35 269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Видин</th>\n",
       "      <td>81 212</td>\n",
       "      <td>39 487</td>\n",
       "      <td>41 725</td>\n",
       "      <td>52 439</td>\n",
       "      <td>25 261</td>\n",
       "      <td>27 178</td>\n",
       "      <td>28 773</td>\n",
       "      <td>14 226</td>\n",
       "      <td>14 547</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   Общо                   В градовете  \\\n",
       "                                 всичко     мъже     жени      всичко   \n",
       "(ОбластиОбщини, ОбластиОбщини)                                          \n",
       "Благоевград                     301 138  146 369  154 769     181 513   \n",
       "Бургас                          409 750  197 692  212 058     311 247   \n",
       "Варна                           470 124  228 625  241 499     390 916   \n",
       "Велико Търново                  229 718  110 803  118 915     160 186   \n",
       "Видин                            81 212   39 487   41 725      52 439   \n",
       "\n",
       "                                                 В селата                  \n",
       "                                   мъже     жени   всичко    мъже    жени  \n",
       "(ОбластиОбщини, ОбластиОбщини)                                             \n",
       "Благоевград                      86 646   94 867  119 625  59 723  59 902  \n",
       "Бургас                          148 393  162 854   98 503  49 299  49 204  \n",
       "Варна                           188 931  201 985   79 208  39 694  39 514  \n",
       "Велико Търново                   76 540   83 646   69 532  34 263  35 269  \n",
       "Видин                            25 261   27 178   28 773  14 226  14 547  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Overall population by region\n",
    "population_soup = BeautifulSoup(requests.get('https://www.nsi.bg/bg/content/2975/население-по-области-общини-местоживеене-и-пол').text)\n",
    "# The HTML table contains both regions and cities, and the index contains duplicate entries for region and the region capitals\n",
    "population_df_full = pd.read_html(str(population_soup.find('table')))[0] \\\n",
    "    .drop_duplicates(subset=[('ОбластиОбщини', 'ОбластиОбщини')], keep='first') \\\n",
    "    .set_index(('ОбластиОбщини', 'ОбластиОбщини'))\n",
    "\n",
    "population_df = population_df_full[population_df_full.index.isin(regions)].sort_index()\n",
    "population_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6c81587c-63dc-4b3a-a9b0-101b7b91fd4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>population</th>\n",
       "      <th>vaccinated</th>\n",
       "      <th>deceased</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Благоевград</th>\n",
       "      <td>301138</td>\n",
       "      <td>72453</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Бургас</th>\n",
       "      <td>409750</td>\n",
       "      <td>120442</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Варна</th>\n",
       "      <td>470124</td>\n",
       "      <td>147092</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Велико Търново</th>\n",
       "      <td>229718</td>\n",
       "      <td>53585</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Видин</th>\n",
       "      <td>81212</td>\n",
       "      <td>17160</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Враца</th>\n",
       "      <td>157637</td>\n",
       "      <td>46901</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Габрово</th>\n",
       "      <td>105788</td>\n",
       "      <td>31928</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Добрич</th>\n",
       "      <td>170298</td>\n",
       "      <td>45791</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Кърджали</th>\n",
       "      <td>160781</td>\n",
       "      <td>47792</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Кюстендил</th>\n",
       "      <td>116619</td>\n",
       "      <td>35000</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ловеч</th>\n",
       "      <td>122490</td>\n",
       "      <td>31004</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Монтана</th>\n",
       "      <td>125395</td>\n",
       "      <td>24507</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Пазарджик</th>\n",
       "      <td>251300</td>\n",
       "      <td>55184</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Перник</th>\n",
       "      <td>120426</td>\n",
       "      <td>29459</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Плевен</th>\n",
       "      <td>233438</td>\n",
       "      <td>67468</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Пловдив</th>\n",
       "      <td>666398</td>\n",
       "      <td>198025</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Разград</th>\n",
       "      <td>109810</td>\n",
       "      <td>29143</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Русе</th>\n",
       "      <td>212729</td>\n",
       "      <td>57919</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Силистра</th>\n",
       "      <td>106852</td>\n",
       "      <td>24939</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Сливен</th>\n",
       "      <td>182551</td>\n",
       "      <td>44391</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Смолян</th>\n",
       "      <td>101887</td>\n",
       "      <td>20960</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>София</th>\n",
       "      <td>238476</td>\n",
       "      <td>77836</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>София (столица)</th>\n",
       "      <td>1308412</td>\n",
       "      <td>554179</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Стара Загора</th>\n",
       "      <td>311400</td>\n",
       "      <td>73019</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Търговище</th>\n",
       "      <td>110027</td>\n",
       "      <td>22573</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Хасково</th>\n",
       "      <td>223625</td>\n",
       "      <td>50483</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Шумен</th>\n",
       "      <td>171781</td>\n",
       "      <td>40695</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ямбол</th>\n",
       "      <td>116486</td>\n",
       "      <td>28932</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 population  vaccinated  deceased\n",
       "Благоевград          301138       72453        31\n",
       "Бургас               409750      120442        43\n",
       "Варна                470124      147092        58\n",
       "Велико Търново       229718       53585        37\n",
       "Видин                 81212       17160        10\n",
       "Враца                157637       46901        14\n",
       "Габрово              105788       31928        14\n",
       "Добрич               170298       45791        23\n",
       "Кърджали             160781       47792        15\n",
       "Кюстендил            116619       35000        17\n",
       "Ловеч                122490       31004        16\n",
       "Монтана              125395       24507        27\n",
       "Пазарджик            251300       55184        36\n",
       "Перник               120426       29459        18\n",
       "Плевен               233438       67468        19\n",
       "Пловдив              666398      198025        82\n",
       "Разград              109810       29143        17\n",
       "Русе                 212729       57919        36\n",
       "Силистра             106852       24939         9\n",
       "Сливен               182551       44391        20\n",
       "Смолян               101887       20960        14\n",
       "София                238476       77836        20\n",
       "София (столица)     1308412      554179        83\n",
       "Стара Загора         311400       73019        48\n",
       "Търговище            110027       22573        17\n",
       "Хасково              223625       50483        41\n",
       "Шумен                171781       40695        30\n",
       "Ямбол                116486       28932        18"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regional_df = pd.DataFrame(dict(\n",
    "    population=pd.to_numeric(population_df[('Общо', 'всичко')].str.replace('\\xa0', '')),\n",
    "    vaccinated=vaccine_df[('Общ брой лица със завършен ваксинационен курс', 'Общ брой лица със завършен ваксинационен курс')],\n",
    "    deceased=deceased_df['14d']\n",
    "))\n",
    "\n",
    "regional_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "01352c6e-c475-43a7-ad6a-f24caefc19bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "78"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Deceased vaccinated by age groups and by 1, 7, 14 day periods\n",
    "deceased_vaccinated_json = requests.get('https://coronavirus.bg/statistics/deceased-vaccinated-7-14.json').json()\n",
    "deceased_vaccinated_df = pd.DataFrame.from_records(deceased_vaccinated_json['vaccinated']).transpose().sum().astype(int)  # Sum by age groups\n",
    "overall_deceased_vaccinated = deceased_vaccinated_df['14']\n",
    "overall_deceased_vaccinated"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "835bb186-b515-48c5-9bc2-b7dba9e9babe",
   "metadata": {},
   "source": [
    "### Historical data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "548daf49-d474-4b1a-a1e4-8284fc60a664",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ваксина</th>\n",
       "      <th>Пол</th>\n",
       "      <th>Възрастова група</th>\n",
       "      <th>Брой починали</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-03-04</th>\n",
       "      <td>COM</td>\n",
       "      <td>Жена</td>\n",
       "      <td>50 - 59</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-04</th>\n",
       "      <td>COM</td>\n",
       "      <td>Жена</td>\n",
       "      <td>60 - 69</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-05</th>\n",
       "      <td>COM</td>\n",
       "      <td>Мъж</td>\n",
       "      <td>50 - 59</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-05</th>\n",
       "      <td>COM</td>\n",
       "      <td>Жена</td>\n",
       "      <td>80 - 89</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-05</th>\n",
       "      <td>JANSS</td>\n",
       "      <td>Мъж</td>\n",
       "      <td>70 - 79</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Ваксина   Пол Възрастова група  Брой починали\n",
       "2022-03-04     COM  Жена          50 - 59              1\n",
       "2022-03-04     COM  Жена          60 - 69              1\n",
       "2022-03-05     COM   Мъж          50 - 59              1\n",
       "2022-03-05     COM  Жена          80 - 89              1\n",
       "2022-03-05   JANSS   Мъж          70 - 79              1"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vaccinated_deceased_df = pd.read_csv(\n",
    "    io.StringIO(requests.get('https://data.egov.bg/resource/download/e6a72183-28e0-486a-b4e4-b5db8b60a900/csv').text),\n",
    "    parse_dates=[0]\n",
    ") \\\n",
    "    .set_index('Дата') \\\n",
    "    .sort_index() \\\n",
    "    .dropna() \\\n",
    "    .rename_axis(index=None)\n",
    "\n",
    "vaccinated_deceased_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "75302a7a-4d7e-4f9b-bb98-df4370a714c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Направени тестове</th>\n",
       "      <th>Тестове за денонощие</th>\n",
       "      <th>Потвърдени случаи</th>\n",
       "      <th>Активни случаи</th>\n",
       "      <th>Нови случаи за денонощие</th>\n",
       "      <th>Хоспитализирани</th>\n",
       "      <th>Новохоспитализирани</th>\n",
       "      <th>В интензивно отделение</th>\n",
       "      <th>Излекувани</th>\n",
       "      <th>Излекувани за денонощие</th>\n",
       "      <th>Починали</th>\n",
       "      <th>Починали за денонощие</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-03-02</th>\n",
       "      <td>9206636</td>\n",
       "      <td>18296</td>\n",
       "      <td>1093920</td>\n",
       "      <td>214709</td>\n",
       "      <td>2641</td>\n",
       "      <td>3652</td>\n",
       "      <td>321</td>\n",
       "      <td>461</td>\n",
       "      <td>843574</td>\n",
       "      <td>3397</td>\n",
       "      <td>35637</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-03</th>\n",
       "      <td>9225405</td>\n",
       "      <td>18771</td>\n",
       "      <td>1096194</td>\n",
       "      <td>212472</td>\n",
       "      <td>2274</td>\n",
       "      <td>3460</td>\n",
       "      <td>330</td>\n",
       "      <td>447</td>\n",
       "      <td>848026</td>\n",
       "      <td>4452</td>\n",
       "      <td>35696</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-04</th>\n",
       "      <td>9234039</td>\n",
       "      <td>8634</td>\n",
       "      <td>1097298</td>\n",
       "      <td>212790</td>\n",
       "      <td>1104</td>\n",
       "      <td>3445</td>\n",
       "      <td>112</td>\n",
       "      <td>436</td>\n",
       "      <td>848792</td>\n",
       "      <td>766</td>\n",
       "      <td>35716</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-05</th>\n",
       "      <td>9253599</td>\n",
       "      <td>19561</td>\n",
       "      <td>1099423</td>\n",
       "      <td>211634</td>\n",
       "      <td>2125</td>\n",
       "      <td>3155</td>\n",
       "      <td>351</td>\n",
       "      <td>420</td>\n",
       "      <td>851999</td>\n",
       "      <td>3207</td>\n",
       "      <td>35790</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-06</th>\n",
       "      <td>9264285</td>\n",
       "      <td>10687</td>\n",
       "      <td>1100811</td>\n",
       "      <td>211620</td>\n",
       "      <td>1388</td>\n",
       "      <td>3175</td>\n",
       "      <td>163</td>\n",
       "      <td>416</td>\n",
       "      <td>853380</td>\n",
       "      <td>1381</td>\n",
       "      <td>35811</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Направени тестове  Тестове за денонощие  Потвърдени случаи  \\\n",
       "2022-03-02            9206636                 18296            1093920   \n",
       "2022-03-03            9225405                 18771            1096194   \n",
       "2022-03-04            9234039                  8634            1097298   \n",
       "2022-03-05            9253599                 19561            1099423   \n",
       "2022-03-06            9264285                 10687            1100811   \n",
       "\n",
       "            Активни случаи  Нови случаи за денонощие  Хоспитализирани  \\\n",
       "2022-03-02          214709                      2641             3652   \n",
       "2022-03-03          212472                      2274             3460   \n",
       "2022-03-04          212790                      1104             3445   \n",
       "2022-03-05          211634                      2125             3155   \n",
       "2022-03-06          211620                      1388             3175   \n",
       "\n",
       "            Новохоспитализирани  В интензивно отделение  Излекувани  \\\n",
       "2022-03-02                  321                     461      843574   \n",
       "2022-03-03                  330                     447      848026   \n",
       "2022-03-04                  112                     436      848792   \n",
       "2022-03-05                  351                     420      851999   \n",
       "2022-03-06                  163                     416      853380   \n",
       "\n",
       "            Излекувани за денонощие  Починали  Починали за денонощие  \n",
       "2022-03-02                     3397     35637                     56  \n",
       "2022-03-03                     4452     35696                     59  \n",
       "2022-03-04                      766     35716                     20  \n",
       "2022-03-05                     3207     35790                     74  \n",
       "2022-03-06                     1381     35811                     21  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history_summary_df = pd.read_csv(\n",
    "    io.StringIO(requests.get('https://data.egov.bg/resource/download/e59f95dd-afde-43af-83c8-ea2916badd19/csv').text),\n",
    "    parse_dates=[0]\n",
    ") \\\n",
    "    .set_index('Дата') \\\n",
    "    .sort_index() \\\n",
    "    .rename_axis(index=None)\n",
    "\n",
    "history_summary_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4105ef26-56e7-4d8f-92f0-8ced804b1891",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>deceased</th>\n",
       "      <th>deceased_vaccinated</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-03-01</th>\n",
       "      <td>99</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-02</th>\n",
       "      <td>56</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-03</th>\n",
       "      <td>59</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-04</th>\n",
       "      <td>20</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-05</th>\n",
       "      <td>74</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            deceased  deceased_vaccinated\n",
       "2022-03-01        99                    4\n",
       "2022-03-02        56                    4\n",
       "2022-03-03        59                    1\n",
       "2022-03-04        20                    6\n",
       "2022-03-05        74                    3"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_history_df = pd.DataFrame(dict(\n",
    "    deceased=history_summary_df['Починали за денонощие'],\n",
    "    deceased_vaccinated=vaccinated_deceased_df['Брой починали'].groupby(vaccinated_deceased_df.index).sum()\n",
    ")) \\\n",
    "    .dropna() \\\n",
    "    .astype(int)\n",
    "\n",
    "full_history_df.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b583f089-c9cf-41c5-9dfc-a9cd55975c5f",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Efficiency estimates based on historical data\n",
    "\n",
    "We take the last month of data and smoothen the curves by first processing individual levels in the Haar wavelet decomposition and then using LOWESS.\n",
    "\n",
    "We then calculate the efficiency directly using the latest population data. Relevant historical population data, for example the total percentage of vaccinated people, is not available, and thus we cannot even calculate efficiency for long-term history.\n",
    "\n",
    "The main goal is to reveal trends and estimate confidence intervals, which we do nonparametrically via bootstrapping."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2b06b16a-cccf-4b63-be35-dc86d614e857",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_days = 4 * 7\n",
    "confidence = 0.95\n",
    "\n",
    "\n",
    "def wavelet_smoothing(array: pd.Series, wavelet='haar') -> pd.Series:\n",
    "    dec = pywt.wavedec(array, wavelet=wavelet)\n",
    "\n",
    "    for i in range(len(dec)):\n",
    "        dec[-i] = pywt.threshold(dec[-i], np.std(dec[-i]), 'soft')\n",
    "\n",
    "    return pd.Series(\n",
    "        pywt.waverec(dec, wavelet=wavelet),\n",
    "        index=array.index\n",
    "    )\n",
    "\n",
    "\n",
    "def lowess_with_ci(array: pd.Series) -> pd.Series:\n",
    "    x = np.arange(len(array))\n",
    "    curve = lowess(\n",
    "        array,\n",
    "        x,\n",
    "        return_sorted=False\n",
    "    )\n",
    "    \n",
    "    samples = [np.random.choice(x, len(array) // 2, replace=False) for i in range(1000)]\n",
    "    randomized_curves = np.array([\n",
    "        scipy.interpolate.interp1d(\n",
    "            fill_value='extrapolate',\n",
    "            x=sam,\n",
    "            y=lowess(\n",
    "                array.iloc[sam],\n",
    "                sam,\n",
    "                return_sorted=False\n",
    "            )\n",
    "        )(x)\n",
    "        for sam in samples\n",
    "    ])\n",
    "    \n",
    "    high = np.percentile(randomized_curves, (1 + confidence) * 50, axis=0)\n",
    "    low = np.percentile(randomized_curves, (1 - confidence) * 50, axis=0)\n",
    "    \n",
    "    return pd.DataFrame(\n",
    "        index=array.index,\n",
    "        data=dict(\n",
    "            mean=curve,\n",
    "            low=low,\n",
    "            high=high\n",
    "        )\n",
    "    )\n",
    "\n",
    "\n",
    "history_df_cleaned = pd.DataFrame(dict(\n",
    "    deceased=wavelet_smoothing(full_history_df['deceased'].tail(n_days)),\n",
    "    deceased_vaccinated=wavelet_smoothing(full_history_df['deceased_vaccinated'].tail(n_days))\n",
    "))\n",
    "\n",
    "deceased_lowess_ci = lowess_with_ci(history_df_cleaned['deceased'])\n",
    "deceased_vaccinated_lowess_ci = lowess_with_ci(history_df_cleaned['deceased_vaccinated'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7e15b165-4c98-43c1-b60f-8ba89178eb8b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "history_comparison_plot_df = pd.concat([\n",
    "    pd.DataFrame(dict(\n",
    "        deceased=full_history_df['deceased'].tail(n_days),\n",
    "        wave='Source',\n",
    "        kind='Deceased'\n",
    "    )),\n",
    "    pd.DataFrame(dict(\n",
    "        deceased=history_df_cleaned['deceased'],\n",
    "        wave='Wavelet filter',\n",
    "        kind='Deceased'\n",
    "    )),\n",
    "    pd.DataFrame(dict(\n",
    "        deceased=deceased_lowess_ci['mean'],\n",
    "        err_lower=deceased_lowess_ci['low'],\n",
    "        err_upper=deceased_lowess_ci['high'],\n",
    "        wave='LOWESS',\n",
    "        kind='Deceased'\n",
    "    )),\n",
    "    pd.DataFrame(dict(\n",
    "        deceased=full_history_df['deceased_vaccinated'].tail(n_days),\n",
    "        wave='Source',\n",
    "        kind='Deceased vaccinated'\n",
    "    )),\n",
    "    pd.DataFrame(dict(\n",
    "        deceased=history_df_cleaned['deceased_vaccinated'],\n",
    "        wave='Wavelet filter',\n",
    "        kind='Deceased vaccinated'\n",
    "    )),\n",
    "    pd.DataFrame(dict(\n",
    "        deceased=deceased_vaccinated_lowess_ci['mean'],\n",
    "        err_lower=deceased_vaccinated_lowess_ci['low'],\n",
    "        err_upper=deceased_vaccinated_lowess_ci['high'],\n",
    "        wave='LOWESS',\n",
    "        kind='Deceased vaccinated'\n",
    "    ))\n",
    "])\n",
    "\n",
    "history_comparison_plot = p9.ggplot(history_comparison_plot_df) \\\n",
    "    + p9.aes(x=history_comparison_plot_df.index, y='deceased', color='wave') \\\n",
    "    + p9.ggtitle('Noise cleaning and 95% CI for process expectations') \\\n",
    "    + p9.labs(y='', fill='', color='') \\\n",
    "    + p9.geom_line() \\\n",
    "    + p9.geom_ribbon(p9.aes(ymin='err_lower', ymax='err_upper', fill='wave'), alpha=0.2, color=None) \\\n",
    "    + p9.scale_x_datetime(date_breaks='1 week') \\\n",
    "    + p9.facet_wrap('kind', scales='free') \\\n",
    "    + p9.theme(figure_size=(10, 3))\n",
    "\n",
    "with warnings.catch_warnings():\n",
    "    warnings.simplefilter('ignore', p9.exceptions.PlotnineWarning)\n",
    "    warnings.simplefilter('ignore', FutureWarning)\n",
    "    history_comparison_plot.draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "22ca267a-19fa-4e8a-a78c-3c3d6746fb86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def apply_fun_to_ci(cis: List[pd.DataFrame], fun: Callable):\n",
    "    evaluations = np.array([\n",
    "        [fun(*t) for t in itertools.product(*(np.linspace(ci['low'][i], ci['high'][i], 10) for ci in cis))]\n",
    "        for i in range(len(cis[0]))\n",
    "    ])\n",
    "    \n",
    "    low = np.min(evaluations, axis=1)\n",
    "    high = np.max(evaluations, axis=1)\n",
    "\n",
    "    return pd.DataFrame(dict(\n",
    "        low=low,\n",
    "        high=high,\n",
    "        mean=fun(*(ci['mean'] for ci in cis))\n",
    "    ))\n",
    "\n",
    "\n",
    "history_percent_deceased_vaccinated_ci = apply_fun_to_ci(\n",
    "    [deceased_vaccinated_lowess_ci],\n",
    "    lambda dv: 100 * dv / regional_df['vaccinated'].sum(),\n",
    ")\n",
    "\n",
    "history_percent_deceased_unvaccinated_ci = apply_fun_to_ci(\n",
    "    [deceased_lowess_ci, deceased_vaccinated_lowess_ci],\n",
    "    lambda d, dv: 100 * (d - dv) / (regional_df['population'].sum() - regional_df['vaccinated'].sum())\n",
    ")\n",
    "\n",
    "history_efficiency_ci = apply_fun_to_ci(\n",
    "    [history_percent_deceased_vaccinated_ci, history_percent_deceased_unvaccinated_ci],\n",
    "    lambda pv, pu: 100 - 100 * pv / pu\n",
    ")\n",
    "\n",
    "history_efficiency_plot_df = pd.concat([\n",
    "    pd.DataFrame(dict(\n",
    "        percent=history_percent_deceased_vaccinated_ci['mean'],\n",
    "        err_lower=history_percent_deceased_vaccinated_ci['low'],\n",
    "        err_upper=history_percent_deceased_vaccinated_ci['high'],\n",
    "        kind='% deceased among vaccinated'\n",
    "    )).tail(n_days),  # Remove the additional points that were added to aid bootstrapping\n",
    "    pd.DataFrame(dict(\n",
    "        percent=history_percent_deceased_unvaccinated_ci['mean'],\n",
    "        err_lower=history_percent_deceased_unvaccinated_ci['low'],\n",
    "        err_upper=history_percent_deceased_unvaccinated_ci['high'],\n",
    "        kind='% deceased among unvaccinated'\n",
    "    )).tail(n_days),\n",
    "    pd.DataFrame(dict(\n",
    "        percent=history_efficiency_ci['mean'],\n",
    "        err_lower=history_efficiency_ci['low'],\n",
    "        err_upper=history_efficiency_ci['high'],\n",
    "        kind='% efficiency'\n",
    "    )).tail(n_days)\n",
    "])\n",
    "\n",
    "history_efficiency_plot = p9.ggplot(history_efficiency_plot_df) \\\n",
    "    + p9.aes(x=history_efficiency_plot_df.index, y='percent') \\\n",
    "    + p9.ggtitle('95% CI for the expectations of derived time series') \\\n",
    "    + p9.ylab('') \\\n",
    "    + p9.geom_line() \\\n",
    "    + p9.geom_ribbon(p9.aes(ymin='err_lower', ymax='err_upper'), alpha=0.2) \\\n",
    "    + p9.scale_x_datetime(date_breaks='1 week') \\\n",
    "    + p9.facet_wrap('kind', scales='free', ncol=2) \\\n",
    "    + p9.theme(figure_size=(10, 6))\n",
    "\n",
    "with warnings.catch_warnings():\n",
    "    warnings.simplefilter('ignore', p9.exceptions.PlotnineWarning)\n",
    "    warnings.simplefilter('ignore', FutureWarning)\n",
    "    history_efficiency_plot.draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f747e91b-c321-40ea-84f9-9ae1a7b6fabd",
   "metadata": {
    "lines_to_next_cell": 2
   },
   "source": [
    "# Efficiency estimates based on regional distribution\n",
    "\n",
    "The main thing to note here is that we do not have information about the deceased vaccinated per region. We add this as an artificial latent variable, and we should take the distribution with a grain of salt, even if the approach and result look plausible. We are only interested in the empirical efficiency distribution. The heuristic assignment of probabilities to regions is more questionable; we only use it as a stepping stone and adjust it so that the efficiency distribution is well-behaved.\n",
    "\n",
    "What we do have is information about the total amount of deceased vaccinated in the last 14 days. We distribute them into different regions iteratively by choosing, at each step, a region for which the standard deviation of the regional efficiencies is minimized. Implementation details can be found a few cells below. We obtain some (perhaps nonlinear) projection of the actual variable.\n",
    "\n",
    "I have previously tried to use principal factors for estimating the deceased vaccinated per region. Unfortunately, that approach often gave a lot of outliers (you can see the details in [this revision](https://gist.github.com/v--/a97892c94959aaf5ae0f7bfa4c026bbd/c7b82d255f76f6a1e1595ac2830664f5eafb778a))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6c168526-57db-4346-a6f5-37dcf47e43ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CI low</th>\n",
       "      <th>CI high</th>\n",
       "      <th>overall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>percent_vaccinated</th>\n",
       "      <td>24.835276</td>\n",
       "      <td>28.436574</td>\n",
       "      <td>29.622581</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       CI low    CI high    overall\n",
       "percent_vaccinated  24.835276  28.436574  29.622581"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We begin with a slight detour\n",
    "# It may happen that the overall efficiency does not fall into the 95% CI for the mean of the regional efficiencies\n",
    "# This is clear from a theoretical standpoint, but for the sake of demonstration we give another example where this is likely to happen without any latent variables\n",
    "overall_percent_vaccinated = 100 * regional_df['vaccinated'].sum() / regional_df['population'].sum()\n",
    "percent_vaccinated = 100 * regional_df['vaccinated'] / regional_df['population']\n",
    "percent_vaccinated_ci = scipy.stats.bootstrap([percent_vaccinated], statistic=np.mean, confidence_level=confidence).confidence_interval\n",
    "\n",
    "pd.DataFrame(index=['percent_vaccinated'], data=[{\n",
    "    'CI low': percent_vaccinated_ci.low,\n",
    "    'CI high': percent_vaccinated_ci.high,\n",
    "    'overall': overall_percent_vaccinated\n",
    "}])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b2dee4c4-02bd-4e0a-b363-8b2ce97e5cc4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8772541425628)>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p9.ggplot(regional_df) \\\n",
    "    + p9.aes(x=percent_vaccinated) \\\n",
    "    + p9.labs(x='% vaccinated', y='Probability density') \\\n",
    "    + p9.geom_density(fill='white') \\\n",
    "    + p9.geom_rect(alpha=0.02, xmin=percent_vaccinated_ci.low, xmax=percent_vaccinated_ci.high, ymin=0, ymax=len(regions)) \\\n",
    "    + p9.geom_vline(xintercept=overall_percent_vaccinated) \\\n",
    "    + p9.annotate('text', label=' 95% CI', x=(percent_vaccinated_ci.low + percent_vaccinated_ci.high) / 2, y=0, va='bottom') \\\n",
    "    + p9.annotate('text', label=' Overall %', x=overall_percent_vaccinated, y=0, ha='left', va='top') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "79b8e8a6-d0ce-4f60-983e-05cd8a921ff8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Now to the essence\n",
    "# This is the algorithm used for determining the number of deceased among the vaccinated\n",
    "deceased_vaccinated = pd.Series(index=regional_df.index, data=0)\n",
    "\n",
    "while deceased_vaccinated.sum() < overall_deceased_vaccinated:\n",
    "    test_series = pd.Series(index=regions, dtype=np.float64)\n",
    "    \n",
    "    for region in regions:\n",
    "        dv = deceased_vaccinated.copy()\n",
    "        dv[region] += 1\n",
    "        ratio_deceased_vaccinated = dv / regional_df['vaccinated']\n",
    "        ratio_deceased_unvaccinated = (regional_df['deceased'] - dv) / (regional_df['population'] - regional_df['vaccinated'])\n",
    "        ratio_efficiency = 1 - ratio_deceased_vaccinated / ratio_deceased_unvaccinated\n",
    "        test_series[region] = ratio_efficiency.std()\n",
    "\n",
    "    deceased_vaccinated[test_series.argmin()] += 1\n",
    "\n",
    "percent_deceased_vaccinated = 100 * deceased_vaccinated / regional_df['vaccinated']\n",
    "percent_deceased_unvaccinated = 100 * (regional_df['deceased'] - deceased_vaccinated) / (regional_df['population'] - regional_df['vaccinated'])\n",
    "efficiency = 100 - 100 * percent_deceased_vaccinated / percent_deceased_unvaccinated"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c9fa17ee-e5ec-46be-9504-7c641e3452c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>WLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>    <td>deceased_vaccinated</td> <th>  R-squared:         </th> <td>   0.987</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                    <td>WLS</td>         <th>  Adj. R-squared:    </th> <td>   0.986</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>              <td>Least Squares</td>    <th>  F-statistic:       </th> <td>   619.6</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>              <td>Sun, 06 Mar 2022</td>   <th>  Prob (F-statistic):</th> <td>7.38e-23</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                  <td>10:56:43</td>       <th>  Log-Likelihood:    </th> <td> -8.6595</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>       <td>    28</td>        <th>  AIC:               </th> <td>   25.32</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>           <td>    24</td>        <th>  BIC:               </th> <td>   30.65</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>               <td>     3</td>        <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>       <td>nonrobust</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "       <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>  <td>   -0.3380</td> <td>    0.147</td> <td>   -2.297</td> <td> 0.031</td> <td>   -0.642</td> <td>   -0.034</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>deceased</th>   <td>    0.0811</td> <td>    0.013</td> <td>    6.078</td> <td> 0.000</td> <td>    0.054</td> <td>    0.109</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>vaccinated</th> <td> 2.751e-05</td> <td> 7.14e-06</td> <td>    3.851</td> <td> 0.001</td> <td> 1.28e-05</td> <td> 4.23e-05</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>population</th> <td>-5.042e-06</td> <td> 3.76e-06</td> <td>   -1.340</td> <td> 0.193</td> <td>-1.28e-05</td> <td> 2.72e-06</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 1.230</td> <th>  Durbin-Watson:     </th> <td>   1.356</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.541</td> <th>  Jarque-Bera (JB):  </th> <td>   1.164</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.379</td> <th>  Prob(JB):          </th> <td>   0.559</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.349</td> <th>  Cond. No.          </th> <td>8.00e+05</td>\n",
       "</tr>\n",
       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large,  8e+05. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                             WLS Regression Results                            \n",
       "===============================================================================\n",
       "Dep. Variable:     deceased_vaccinated   R-squared:                       0.987\n",
       "Model:                             WLS   Adj. R-squared:                  0.986\n",
       "Method:                  Least Squares   F-statistic:                     619.6\n",
       "Date:                 Sun, 06 Mar 2022   Prob (F-statistic):           7.38e-23\n",
       "Time:                         10:56:43   Log-Likelihood:                -8.6595\n",
       "No. Observations:                   28   AIC:                             25.32\n",
       "Df Residuals:                       24   BIC:                             30.65\n",
       "Df Model:                            3                                         \n",
       "Covariance Type:             nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Intercept     -0.3380      0.147     -2.297      0.031      -0.642      -0.034\n",
       "deceased       0.0811      0.013      6.078      0.000       0.054       0.109\n",
       "vaccinated  2.751e-05   7.14e-06      3.851      0.001    1.28e-05    4.23e-05\n",
       "population -5.042e-06   3.76e-06     -1.340      0.193   -1.28e-05    2.72e-06\n",
       "==============================================================================\n",
       "Omnibus:                        1.230   Durbin-Watson:                   1.356\n",
       "Prob(Omnibus):                  0.541   Jarque-Bera (JB):                1.164\n",
       "Skew:                           0.379   Prob(JB):                        0.559\n",
       "Kurtosis:                       2.349   Cond. No.                     8.00e+05\n",
       "==============================================================================\n",
       "\n",
       "Notes:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large,  8e+05. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = wls(\n",
    "    'deceased_vaccinated ~ deceased + vaccinated + population',\n",
    "    data=regional_df.merge(pd.DataFrame(dict(deceased_vaccinated=deceased_vaccinated)), on=regions).set_index('key_0')\n",
    ")\n",
    "\n",
    "model.fit().summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "bb56ba86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8772539693660)>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "overall_percent_deceased_vaccinated = 100 * overall_deceased_vaccinated / regional_df['vaccinated'].sum()\n",
    "overall_percent_deceased_unvaccinated = 100 * (regional_df['deceased'].sum() - overall_deceased_vaccinated) / (regional_df['population'].sum() - regional_df['vaccinated'].sum())\n",
    "\n",
    "vaccinated_percent_plot_df = pd.concat([\n",
    "    pd.DataFrame(dict(\n",
    "        percent=percent_deceased_vaccinated,\n",
    "        kind='Vaccinated',\n",
    "        overall_percent=overall_percent_deceased_vaccinated\n",
    "    )),\n",
    "    pd.DataFrame(dict(\n",
    "        percent=percent_deceased_unvaccinated,\n",
    "        kind='Unvaccinated',\n",
    "        overall_percent=overall_percent_deceased_unvaccinated\n",
    "    ))\n",
    "])\n",
    "\n",
    "p9.ggplot(vaccinated_percent_plot_df) \\\n",
    "    + p9.aes(x='percent', fill='kind', color='kind') \\\n",
    "    + p9.ggtitle('Density estimation of mortality percentages') \\\n",
    "    + p9.labs(x='Percent deceased among (un)vaccinated', y='Probability density', color='', fill='') \\\n",
    "    + p9.geom_density(alpha=0.2) \\\n",
    "    + p9.geom_vline(mapping=p9.aes(xintercept='overall_percent')) \\\n",
    "    + p9.annotate('text', label=' Overall %', ha='left', x=overall_percent_deceased_vaccinated, y=0, va='bottom') \\\n",
    "    + p9.annotate('text', label=' Overall %', ha='left', x=overall_percent_deceased_unvaccinated, y=0, va='bottom')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a995c93f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CI low</th>\n",
       "      <th>CI high</th>\n",
       "      <th>overall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>efficiency</th>\n",
       "      <td>70.447147</td>\n",
       "      <td>75.438151</td>\n",
       "      <td>74.787395</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               CI low    CI high    overall\n",
       "efficiency  70.447147  75.438151  74.787395"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "efficiency_ci = scipy.stats.bootstrap([efficiency], statistic=np.mean, confidence_level=confidence).confidence_interval\n",
    "efficiency_ci_center = (efficiency_ci.high + efficiency_ci.low) / 2\n",
    "overall_efficiency = 100 - 100 * overall_percent_deceased_vaccinated / overall_percent_deceased_unvaccinated\n",
    "\n",
    "pd.DataFrame(index=['efficiency'], data=[{\n",
    "    'CI low': efficiency_ci.low,\n",
    "    'CI high': efficiency_ci.high,\n",
    "    'overall': overall_efficiency\n",
    "}])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "91fecbc2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (8772539696062)>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p9.ggplot(regional_df) \\\n",
    "    + p9.aes(x=efficiency) \\\n",
    "    + p9.labs(x='Estimated vaccine efficiency', y='Probability density') \\\n",
    "    + p9.geom_density(fill='white') \\\n",
    "    + p9.geom_rect(alpha=0.02, xmin=efficiency_ci.low, xmax=efficiency_ci.high, ymin=0, ymax=len(regions)) \\\n",
    "    + p9.geom_vline(xintercept=overall_efficiency) \\\n",
    "    + p9.annotate('text', label='95% CI', x=(efficiency_ci.low + efficiency_ci.high) / 2, y=0, va='bottom') \\\n",
    "    + p9.annotate('text', label=' Overall %', x=overall_efficiency, y=0, ha='left', va='top') "
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "encoding": "# -*- coding: utf-8 -*-",
   "formats": "ipynb,py:percent"
  },
  "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": "3.10.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}