Creating_synthetic_dataForProjection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d399698d",
   "metadata": {},
   "source": [
    "#### To project the data for future dates: \n",
    "##### Dates: From 22-Jun-2021 until 01-Mar-2022\n",
    "We project the data from 22-Jun-2021\n",
    "\n",
    "\n",
    "<br>\n",
    "\n",
    "- Vaccination ratio:\n",
    "                     Approach- linear regression model.\n",
    "                     Reason - This has to keep increasing because of all the vaccination drives happening in the world. So the max value for vaccination ratio is 1\n",
    "<br>\n",
    "\n",
    "- Stringency index: \n",
    "                    Approach - ARIMA / linear regression model\n",
    "                    Reason - This can be projected using a regression model as we see that it keeps decreasing gradually over a period of time.\n",
    "<br>\n",
    "\n",
    "- positive rate: \n",
    "                    Approach - ARIMA / RNN model\n",
    "                    Reason - This has got the second highest F-score on the XGboost model and having a good estimate at future point of time(synthetic data) will directly impact our prediction on that future date\n",
    "\n",
    "<br>\n",
    "\n",
    "- reproduction rate: \n",
    "                    Approach - ARIMA / RNN model\n",
    "                    Reason - Since this has the highest F-score, our reproduction rate projections will directly impact our Prediction model. We can go ahead and use an ARIMA/RNN model to synthetically this\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "901e4dc7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import datetime\n",
    "import seaborn as sb\n",
    "import matplotlib.ticker as ticker\n",
    "import scipy.stats as sp\n",
    "from sklearn import linear_model\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report,mean_squared_error, mean_absolute_error\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "from statsmodels.tsa.arima.model import ARIMA\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "from datetime import timedelta\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "import pandas as pd\n",
    "import scipy\n",
    "import math\n",
    "\n",
    "import xgboost as xgb\n",
    "from xgboost import plot_importance, plot_tree\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "58072045",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample_df = pd.read_csv(\"Prepared_data/Global_PreProcessed_dataset_unscaled.csv\")\n",
    "sample_df['date']  = pd.to_datetime(sample_df['date'], format = '%Y-%m-%d')\n",
    "sample_df = sample_df.set_index('date')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5ac402a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "end_date = '01-Mar-2022'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "173bf000",
   "metadata": {},
   "source": [
    "### Synthetically projecting Vaccination ratio\n",
    "\n",
    "- We know that vaccination has started roughly from Jan-2021. We use these dates to fit the linear regression model from Jan-2021 until 01-May-2021\n",
    "- From 01-May-2021 until Jun-23 we can see the models performance on this test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "08013663",
   "metadata": {},
   "outputs": [],
   "source": [
    "start_date = '01-Jan-2021'\n",
    "vaccination_df = sample_df.loc[sample_df.index>=start_date].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d6adf4f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "vaccination_df = vaccination_df.drop(['new_deaths_per_million','reproduction_rate','positive_rate','stringency_index'],axis=1)\n",
    "\n",
    "_ = vaccination_df.plot(figsize=(15,5), title='Vaccination_ratio dataset')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "13ed3ddd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# To have a reference from a start date \n",
    "vaccination_df['days_from_start'] = (vaccination_df.index - vaccination_df.index[0]).days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "480594b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "vaccination_df = vaccination_df.loc[vaccination_df.vaccination_ratio !=0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7dafce95",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "split_date = '01-Jun-2021'\n",
    "vaccination_df_train = vaccination_df.loc[vaccination_df.index <= split_date].copy()\n",
    "vaccination_df_test = vaccination_df.loc[vaccination_df.index > split_date].copy()\n",
    "\n",
    "X_train = vaccination_df_train.drop(['vaccination_ratio'],axis=1)\n",
    "y_train = pd.DataFrame(vaccination_df_train['vaccination_ratio'])\n",
    "\n",
    "X_test = vaccination_df_test.drop(['vaccination_ratio'],axis=1)\n",
    "y_test = pd.DataFrame(vaccination_df_test['vaccination_ratio'])\n",
    "\n",
    "model = linear_model.LinearRegression().fit(X_train, y_train)#\n",
    "vaccination_df_test['vaccination_ratio_Prediction_linreg'] = model.predict(X_test) \n",
    "\n",
    "vaccination_df_all = pd.concat([vaccination_df_train,vaccination_df_test], sort=False)\n",
    "_ = vaccination_df_all[['vaccination_ratio','vaccination_ratio_Prediction_linreg']].plot(figsize=(15, 5),title=\"Prediction of model on vaccination data\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "103035cf",
   "metadata": {},
   "source": [
    "> We see that vaccination ratio is not exactly linear. It is more of exponential. This can b explained by the fact that there are more vaccination drives happening and people are getting vaccinated faster w.r.t time\n",
    "> Alternative is to go ahead to fit the vaccation_ratio using a simple Neural network which can handle curves\n",
    "   - The vaccination ratio has a max_limit of 1, so we can go ahead and use a sigmoid function in our final layer of NN\n",
    "   - The hidden layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "66dbb374",
   "metadata": {},
   "outputs": [],
   "source": [
    "days = pd.date_range(start='2021-06-22', end=end_date)\n",
    "future_projections = pd.DataFrame(data=days, index = range(len(days)), columns=['date'])\n",
    "future_projections = future_projections.set_index('date')\n",
    "#projection_10_months = vaccination_df.append(projection_10_months)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "68b81c39",
   "metadata": {},
   "outputs": [],
   "source": [
    "future_projections['days_from_start'] = (future_projections.index - vaccination_df.index[0]).days\n",
    "visualization_df = future_projections.copy()\n",
    "future_projections['vaccination_ratio'] = model.predict(future_projections) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "fa2e1f90",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We create a separate dataframe so that we can visualize it\n",
    "visualization_df['vaccination_ratio_Prediction'] = model.predict(visualization_df) \n",
    "visualization_df = vaccination_df.append(visualization_df)\n",
    "_ = visualization_df[['vaccination_ratio','vaccination_ratio_Prediction']].plot(figsize=(15, 5),title=\"Prediction of model on vaccination data\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d68992c8",
   "metadata": {},
   "source": [
    "This is our synthetic data, however this can be improved by using a simple neural network to capture the non linearity mentioned previously. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a3889c1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "vaccination_ratio_future_projections = future_projections"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c8b83a7",
   "metadata": {},
   "source": [
    "### Synthetically projecting Stringency Index\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7f86f925",
   "metadata": {},
   "outputs": [],
   "source": [
    "future_projections.insert(loc=2,column='stringency_index',value=[0.0]*len(future_projections))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "37eb9c15",
   "metadata": {},
   "outputs": [],
   "source": [
    "stringency_df = sample_df.drop(sample_df.columns.difference(['stringency_index']), 1, inplace=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c57e770b",
   "metadata": {},
   "outputs": [],
   "source": [
    "stringency_df = stringency_df.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "2196dcfc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing stepwise search to minimize aic\n",
      " ARIMA(1,1,1)(0,1,1)[12]             : AIC=inf, Time=1.15 sec\n",
      " ARIMA(0,1,0)(0,1,0)[12]             : AIC=1363.800, Time=0.03 sec\n",
      " ARIMA(1,1,0)(1,1,0)[12]             : AIC=1191.180, Time=0.18 sec\n",
      " ARIMA(0,1,1)(0,1,1)[12]             : AIC=1123.744, Time=0.39 sec\n",
      " ARIMA(0,1,1)(0,1,0)[12]             : AIC=1326.621, Time=0.05 sec\n",
      " ARIMA(0,1,1)(1,1,1)[12]             : AIC=inf, Time=0.56 sec\n",
      " ARIMA(0,1,1)(0,1,2)[12]             : AIC=inf, Time=1.52 sec\n",
      " ARIMA(0,1,1)(1,1,0)[12]             : AIC=1233.661, Time=0.15 sec\n",
      " ARIMA(0,1,1)(1,1,2)[12]             : AIC=inf, Time=3.08 sec\n",
      " ARIMA(0,1,0)(0,1,1)[12]             : AIC=1212.761, Time=0.21 sec\n",
      " ARIMA(0,1,2)(0,1,1)[12]             : AIC=1105.956, Time=0.49 sec\n",
      " ARIMA(0,1,2)(0,1,0)[12]             : AIC=1316.252, Time=0.10 sec\n",
      " ARIMA(0,1,2)(1,1,1)[12]             : AIC=inf, Time=0.70 sec\n",
      " ARIMA(0,1,2)(0,1,2)[12]             : AIC=inf, Time=1.78 sec\n",
      " ARIMA(0,1,2)(1,1,0)[12]             : AIC=1216.940, Time=0.24 sec\n",
      " ARIMA(0,1,2)(1,1,2)[12]             : AIC=inf, Time=3.33 sec\n",
      " ARIMA(1,1,2)(0,1,1)[12]             : AIC=inf, Time=0.87 sec\n",
      " ARIMA(0,1,3)(0,1,1)[12]             : AIC=970.664, Time=0.96 sec\n",
      " ARIMA(0,1,3)(0,1,0)[12]             : AIC=1196.748, Time=0.16 sec\n",
      " ARIMA(0,1,3)(1,1,1)[12]             : AIC=inf, Time=1.12 sec\n",
      " ARIMA(0,1,3)(0,1,2)[12]             : AIC=inf, Time=2.40 sec\n",
      " ARIMA(0,1,3)(1,1,0)[12]             : AIC=1096.352, Time=0.35 sec\n",
      " ARIMA(0,1,3)(1,1,2)[12]             : AIC=inf, Time=4.02 sec\n",
      " ARIMA(1,1,3)(0,1,1)[12]             : AIC=inf, Time=1.36 sec\n",
      " ARIMA(0,1,3)(0,1,1)[12] intercept   : AIC=inf, Time=1.54 sec\n",
      "\n",
      "Best model:  ARIMA(0,1,3)(0,1,1)[12]          \n",
      "Total fit time: 26.769 seconds\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>                   <td>y</td>                <th>  No. Observations:  </th>    <td>539</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>SARIMAX(0, 1, 3)x(0, 1, [1], 12)</td> <th>  Log Likelihood     </th> <td>-480.332</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                    <td>Sun, 27 Jun 2021</td>         <th>  AIC                </th>  <td>970.664</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                        <td>13:39:58</td>             <th>  BIC                </th>  <td>991.991</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                          <td>0</td>                <th>  HQIC               </th>  <td>979.014</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                              <td> - 539</td>              <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>                <td>opg</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>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>    <td>    0.2539</td> <td>    0.027</td> <td>    9.434</td> <td> 0.000</td> <td>    0.201</td> <td>    0.307</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L2</th>    <td>    0.2631</td> <td>    0.020</td> <td>   13.366</td> <td> 0.000</td> <td>    0.225</td> <td>    0.302</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L3</th>    <td>    0.5204</td> <td>    0.029</td> <td>   17.692</td> <td> 0.000</td> <td>    0.463</td> <td>    0.578</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L12</th> <td>   -0.8807</td> <td>    0.034</td> <td>  -26.190</td> <td> 0.000</td> <td>   -0.947</td> <td>   -0.815</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>   <td>    0.3509</td> <td>    0.010</td> <td>   36.923</td> <td> 0.000</td> <td>    0.332</td> <td>    0.370</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.62</td> <th>  Jarque-Bera (JB):  </th> <td>8609.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.43</td> <th>  Prob(JB):          </th>  <td>0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>0.18</td> <th>  Skew:              </th>  <td>1.36</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.00</td> <th>  Kurtosis:          </th>  <td>22.63</td> \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                      SARIMAX Results                                       \n",
       "============================================================================================\n",
       "Dep. Variable:                                    y   No. Observations:                  539\n",
       "Model:             SARIMAX(0, 1, 3)x(0, 1, [1], 12)   Log Likelihood                -480.332\n",
       "Date:                              Sun, 27 Jun 2021   AIC                            970.664\n",
       "Time:                                      13:39:58   BIC                            991.991\n",
       "Sample:                                           0   HQIC                           979.014\n",
       "                                              - 539                                         \n",
       "Covariance Type:                                opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ma.L1          0.2539      0.027      9.434      0.000       0.201       0.307\n",
       "ma.L2          0.2631      0.020     13.366      0.000       0.225       0.302\n",
       "ma.L3          0.5204      0.029     17.692      0.000       0.463       0.578\n",
       "ma.S.L12      -0.8807      0.034    -26.190      0.000      -0.947      -0.815\n",
       "sigma2         0.3509      0.010     36.923      0.000       0.332       0.370\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   0.62   Jarque-Bera (JB):              8609.00\n",
       "Prob(Q):                              0.43   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               0.18   Skew:                             1.36\n",
       "Prob(H) (two-sided):                  0.00   Kurtosis:                        22.63\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pmdarima import auto_arima\n",
    "  \n",
    "# Ignore harmless warnings\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "\n",
    "## We try to find the best arima model for our data\n",
    "# Fit auto_arima function to our dataset\n",
    "stepwise_fit = auto_arima(stringency_df['stringency_index'], start_p = 1, start_q = 1,\n",
    "                          max_p = 5, max_q = 3, m = 12,\n",
    "                          start_P = 0, seasonal = True,\n",
    "                          d = None, D = 1, trace = True,\n",
    "                          error_action ='ignore',   # we don't want to know if an order does not work\n",
    "                          suppress_warnings = True,  # we don't want convergence warnings\n",
    "                          stepwise = True)           # set to stepwise\n",
    "  \n",
    "# To print the summary\n",
    "stepwise_fit.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "fcd9c271",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "split_date = '01-Jun-2021'\n",
    "stringency_df_train = stringency_df.loc[stringency_df.index <= split_date].copy()\n",
    "stringency_df_test = stringency_df.loc[stringency_df.index > split_date].copy()\n",
    "\n",
    "# (4, 1, 0)x(2, 1, [1], 12)\n",
    "# Fit a SARIMAX(0, 1, 3)x(0, 1, 1, 12) on the training set\n",
    "from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
    "\n",
    "model = SARIMAX(stringency_df_train['stringency_index'],\n",
    "                order = (0, 1, 3), \n",
    "                seasonal_order =(0, 1, [1], 12))\n",
    "\n",
    "result = model.fit()\n",
    "#result.summary()\n",
    "\n",
    "start = len(stringency_df_train)\n",
    "end = len(stringency_df_train) + len(stringency_df_test) - 1\n",
    "  \n",
    "# Predictions against the test set\n",
    "predictions = result.predict(start, end,\n",
    "                             typ = 'levels').rename(\"Predictions\")\n",
    "  \n",
    "# plot predictions and actual values\n",
    "predictions.plot(legend = True)\n",
    "stringency_df_test['stringency_index'].plot(legend = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dee97131",
   "metadata": {},
   "source": [
    "Here we can see that the SARIMAX model has gathered a trend in our data, we now use this model to forecast future stringency_index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "0c543517",
   "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>days_from_start</th>\n",
       "      <th>vaccination_ratio</th>\n",
       "      <th>stringency_index</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2021-06-22</th>\n",
       "      <td>172</td>\n",
       "      <td>0.256474</td>\n",
       "      <td>51.868969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-23</th>\n",
       "      <td>173</td>\n",
       "      <td>0.258015</td>\n",
       "      <td>51.790031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-24</th>\n",
       "      <td>174</td>\n",
       "      <td>0.259557</td>\n",
       "      <td>51.771781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-25</th>\n",
       "      <td>175</td>\n",
       "      <td>0.261098</td>\n",
       "      <td>51.611657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-26</th>\n",
       "      <td>176</td>\n",
       "      <td>0.262640</td>\n",
       "      <td>51.577481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-25</th>\n",
       "      <td>420</td>\n",
       "      <td>0.638754</td>\n",
       "      <td>43.853864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-26</th>\n",
       "      <td>421</td>\n",
       "      <td>0.640295</td>\n",
       "      <td>43.899959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-27</th>\n",
       "      <td>422</td>\n",
       "      <td>0.641837</td>\n",
       "      <td>43.895360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-28</th>\n",
       "      <td>423</td>\n",
       "      <td>0.643378</td>\n",
       "      <td>43.979936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-01</th>\n",
       "      <td>424</td>\n",
       "      <td>0.644920</td>\n",
       "      <td>43.949724</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>253 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            days_from_start  vaccination_ratio  stringency_index\n",
       "date                                                            \n",
       "2021-06-22              172           0.256474         51.868969\n",
       "2021-06-23              173           0.258015         51.790031\n",
       "2021-06-24              174           0.259557         51.771781\n",
       "2021-06-25              175           0.261098         51.611657\n",
       "2021-06-26              176           0.262640         51.577481\n",
       "...                     ...                ...               ...\n",
       "2022-02-25              420           0.638754         43.853864\n",
       "2022-02-26              421           0.640295         43.899959\n",
       "2022-02-27              422           0.641837         43.895360\n",
       "2022-02-28              423           0.643378         43.979936\n",
       "2022-03-01              424           0.644920         43.949724\n",
       "\n",
       "[253 rows x 3 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "start = len(stringency_df)-1\n",
    "end = len(stringency_df)+len(future_projections) - 1\n",
    "  \n",
    "# Predictions against the test set\n",
    "stringency_predictions = result.predict(start, end,\n",
    "                             typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "# plot predictions and actual values\n",
    "stringency_predictions.plot(legend = True)\n",
    "stringency_df['stringency_index'].plot(legend = True)\n",
    "\n",
    "result.predict(start, end,typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "future_projections['stringency_index'] = stringency_predictions\n",
    "future_projections"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92b092ea",
   "metadata": {},
   "source": [
    "- Well this what the future holds for us in terms of Stringency_index"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7011047f",
   "metadata": {},
   "source": [
    "### Synthetically projecting reproduction_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "06c8f444",
   "metadata": {},
   "outputs": [],
   "source": [
    "future_projections.insert(loc=2,column='reproduction_rate',value=[0.0]*len(future_projections))\n",
    "rrate_df = sample_df.drop(sample_df.columns.difference(['reproduction_rate']), 1, inplace=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "26e603ed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing stepwise search to minimize aic\n",
      " ARIMA(1,1,1)(0,1,1)[12]             : AIC=inf, Time=0.84 sec\n",
      " ARIMA(0,1,0)(0,1,0)[12]             : AIC=-595.894, Time=0.10 sec\n",
      " ARIMA(1,1,0)(1,1,0)[12]             : AIC=-776.979, Time=0.35 sec\n",
      " ARIMA(0,1,1)(0,1,1)[12]             : AIC=inf, Time=0.89 sec\n",
      " ARIMA(1,1,0)(0,1,0)[12]             : AIC=-600.839, Time=0.04 sec\n",
      " ARIMA(1,1,0)(2,1,0)[12]             : AIC=-824.068, Time=0.79 sec\n",
      " ARIMA(1,1,0)(2,1,1)[12]             : AIC=inf, Time=2.62 sec\n",
      " ARIMA(1,1,0)(1,1,1)[12]             : AIC=inf, Time=0.78 sec\n",
      " ARIMA(0,1,0)(2,1,0)[12]             : AIC=-824.877, Time=0.55 sec\n",
      " ARIMA(0,1,0)(1,1,0)[12]             : AIC=-775.523, Time=0.17 sec\n",
      " ARIMA(0,1,0)(2,1,1)[12]             : AIC=inf, Time=3.55 sec\n",
      " ARIMA(0,1,0)(1,1,1)[12]             : AIC=inf, Time=0.53 sec\n",
      " ARIMA(0,1,1)(2,1,0)[12]             : AIC=-824.349, Time=0.89 sec\n",
      " ARIMA(1,1,1)(2,1,0)[12]             : AIC=-823.156, Time=1.49 sec\n",
      " ARIMA(0,1,0)(2,1,0)[12] intercept   : AIC=-824.356, Time=1.81 sec\n",
      "\n",
      "Best model:  ARIMA(0,1,0)(2,1,0)[12]          \n",
      "Total fit time: 15.415 seconds\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>                  <td>y</td>               <th>  No. Observations:  </th>    <td>539</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>SARIMAX(0, 1, 0)x(2, 1, 0, 12)</td> <th>  Log Likelihood     </th>  <td>415.438</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                   <td>Sun, 27 Jun 2021</td>        <th>  AIC                </th> <td>-824.877</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                       <td>13:41:03</td>            <th>  BIC                </th> <td>-812.081</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                         <td>0</td>               <th>  HQIC               </th> <td>-819.867</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                             <td> - 539</td>             <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>               <td>opg</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>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.S.L12</th> <td>   -0.7091</td> <td>    0.009</td> <td>  -77.436</td> <td> 0.000</td> <td>   -0.727</td> <td>   -0.691</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.S.L24</th> <td>   -0.3100</td> <td>    0.009</td> <td>  -32.866</td> <td> 0.000</td> <td>   -0.328</td> <td>   -0.291</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>   <td>    0.0119</td> <td> 8.32e-05</td> <td>  143.121</td> <td> 0.000</td> <td>    0.012</td> <td>    0.012</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>1.31</td> <th>  Jarque-Bera (JB):  </th> <td>523067.29</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.25</td> <th>  Prob(JB):          </th>   <td>0.00</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>0.00</td> <th>  Skew:              </th>   <td>8.35</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.00</td> <th>  Kurtosis:          </th>  <td>156.58</td>  \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                     SARIMAX Results                                      \n",
       "==========================================================================================\n",
       "Dep. Variable:                                  y   No. Observations:                  539\n",
       "Model:             SARIMAX(0, 1, 0)x(2, 1, 0, 12)   Log Likelihood                 415.438\n",
       "Date:                            Sun, 27 Jun 2021   AIC                           -824.877\n",
       "Time:                                    13:41:03   BIC                           -812.081\n",
       "Sample:                                         0   HQIC                          -819.867\n",
       "                                            - 539                                         \n",
       "Covariance Type:                              opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ar.S.L12      -0.7091      0.009    -77.436      0.000      -0.727      -0.691\n",
       "ar.S.L24      -0.3100      0.009    -32.866      0.000      -0.328      -0.291\n",
       "sigma2         0.0119   8.32e-05    143.121      0.000       0.012       0.012\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   1.31   Jarque-Bera (JB):            523067.29\n",
       "Prob(Q):                              0.25   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               0.00   Skew:                             8.35\n",
       "Prob(H) (two-sided):                  0.00   Kurtosis:                       156.58\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pmdarima import auto_arima\n",
    "\n",
    "# Ignore harmless warnings\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "\n",
    "## We try to find the best arima model for our data\n",
    "# Fit auto_arima function to our dataset\n",
    "stepwise_fit = auto_arima(rrate_df['reproduction_rate'], start_p = 1, start_q = 1,\n",
    "                          max_p = 5, max_q = 3, m = 12,\n",
    "                          start_P = 0, seasonal = True,\n",
    "                          d = None, D = 1, trace = True,\n",
    "                          error_action ='ignore',   # we don't want to know if an order does not work\n",
    "                          suppress_warnings = True,  # we don't want convergence warnings\n",
    "                          stepwise = True)           # set to stepwise\n",
    "  \n",
    "# To print the summary\n",
    "stepwise_fit.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "13800610",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "split_date = '01-Jun-2021'\n",
    "\n",
    "rrate_df_train = rrate_df.loc[rrate_df.index <= split_date].copy()\n",
    "rrate_df_test = rrate_df.loc[rrate_df.index > split_date].copy()\n",
    "\n",
    "# Fit a SARIMAX(0, 1, 0)x(0, 1, [1], 12) on the training set\n",
    "from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
    "\n",
    "model = SARIMAX(rrate_df_train['reproduction_rate'], \n",
    "                order = (0, 1, 0), \n",
    "                seasonal_order =(2, 1, 0, 12))\n",
    "  \n",
    "result = model.fit()\n",
    "#result.summary()\n",
    "\n",
    "start = len(rrate_df_train)\n",
    "end = len(rrate_df_train) + len(rrate_df_test) - 1\n",
    "  \n",
    "# Predictions against the test set\n",
    "predictions = result.predict(start, end,\n",
    "                             typ = 'levels').rename(\"Predictions\")\n",
    "  \n",
    "# plot predictions and actual values\n",
    "predictions.plot(legend = True)\n",
    "rrate_df_test['reproduction_rate'].plot(legend = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb9e987c",
   "metadata": {},
   "source": [
    "The model has gathered an overall trend, however it is far from the actual. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "a5b19b97",
   "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>days_from_start</th>\n",
       "      <th>vaccination_ratio</th>\n",
       "      <th>reproduction_rate</th>\n",
       "      <th>stringency_index</th>\n",
       "      <th>positive_rate</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</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>2021-06-22</th>\n",
       "      <td>172</td>\n",
       "      <td>0.256474</td>\n",
       "      <td>0.886064</td>\n",
       "      <td>51.868969</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-23</th>\n",
       "      <td>173</td>\n",
       "      <td>0.258015</td>\n",
       "      <td>0.887335</td>\n",
       "      <td>51.790031</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-24</th>\n",
       "      <td>174</td>\n",
       "      <td>0.259557</td>\n",
       "      <td>0.888973</td>\n",
       "      <td>51.771781</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-25</th>\n",
       "      <td>175</td>\n",
       "      <td>0.261098</td>\n",
       "      <td>0.889609</td>\n",
       "      <td>51.611657</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-26</th>\n",
       "      <td>176</td>\n",
       "      <td>0.262640</td>\n",
       "      <td>0.888440</td>\n",
       "      <td>51.577481</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-25</th>\n",
       "      <td>420</td>\n",
       "      <td>0.638754</td>\n",
       "      <td>0.806695</td>\n",
       "      <td>43.853864</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-26</th>\n",
       "      <td>421</td>\n",
       "      <td>0.640295</td>\n",
       "      <td>0.806359</td>\n",
       "      <td>43.899959</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-27</th>\n",
       "      <td>422</td>\n",
       "      <td>0.641837</td>\n",
       "      <td>0.806788</td>\n",
       "      <td>43.895360</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-28</th>\n",
       "      <td>423</td>\n",
       "      <td>0.643378</td>\n",
       "      <td>0.807659</td>\n",
       "      <td>43.979936</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-01</th>\n",
       "      <td>424</td>\n",
       "      <td>0.644920</td>\n",
       "      <td>0.808710</td>\n",
       "      <td>43.949724</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>253 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            days_from_start  vaccination_ratio  reproduction_rate  \\\n",
       "date                                                                \n",
       "2021-06-22              172           0.256474           0.886064   \n",
       "2021-06-23              173           0.258015           0.887335   \n",
       "2021-06-24              174           0.259557           0.888973   \n",
       "2021-06-25              175           0.261098           0.889609   \n",
       "2021-06-26              176           0.262640           0.888440   \n",
       "...                     ...                ...                ...   \n",
       "2022-02-25              420           0.638754           0.806695   \n",
       "2022-02-26              421           0.640295           0.806359   \n",
       "2022-02-27              422           0.641837           0.806788   \n",
       "2022-02-28              423           0.643378           0.807659   \n",
       "2022-03-01              424           0.644920           0.808710   \n",
       "\n",
       "            stringency_index  positive_rate  \n",
       "date                                         \n",
       "2021-06-22         51.868969            0.0  \n",
       "2021-06-23         51.790031            0.0  \n",
       "2021-06-24         51.771781            0.0  \n",
       "2021-06-25         51.611657            0.0  \n",
       "2021-06-26         51.577481            0.0  \n",
       "...                      ...            ...  \n",
       "2022-02-25         43.853864            0.0  \n",
       "2022-02-26         43.899959            0.0  \n",
       "2022-02-27         43.895360            0.0  \n",
       "2022-02-28         43.979936            0.0  \n",
       "2022-03-01         43.949724            0.0  \n",
       "\n",
       "[253 rows x 5 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAESCAYAAADUqZ9PAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAAqFklEQVR4nO3dd5xdVb338c9vzpzpJW3SSJmAEAiQBBhCCYQEkFAi7YIRUIgKyPWq8b4eQIEHBO+Dyr0oKnpFRIIoIihSjLRQEzppQEiEUIY0SC9TMu2c9fyxzkwmkyln5rQ9k+/79TqvfWbX35zZ89vrrL3W2uacQ0REgisr0wGIiEjnlKhFRAJOiVpEJOCUqEVEAk6JWkQk4JSoRUQCLjsVOx00aJArLy9Pxa5FRPqkRYsWbXLOlbW3LCWJury8nIULF6Zi1yIifZKZfdLRMlV9iIgEnBK1iEjAKVGLiARcSuqoRaR7GhsbWbNmDXV1dZkORVIsLy+PESNGEA6H495GiVokANasWUNxcTHl5eWYWabDkRRxzrF582bWrFnDmDFj4t5OVR8iAVBXV8fAgQOVpPs4M2PgwIHd/uaUmkTtoinZrUhfpiS9d+jJ3zk1ibq+KiW7FZHUCYVCTJw4kUMOOYTzzz+f2traHu9r1qxZ/O1vfwPg0ksvZfny5R2u+8ILL/DKK6+0/HzHHXdw77339vjYfVFqEnWkMSW7FZHUyc/PZ+nSpSxbtoycnBzuuOOO3ZY3NTX1aL933XUX48aN63B520R9xRVXcPHFF/foWH1VahJ1VIlapDc7/vjj+eCDD3jhhRc4/vjjOfPMMxk3bhyRSISrrrqKI488kvHjx/Pb3/4W8DfJvvWtbzF27FhOPvlkNmzY0LKvqVOntvRUfvLJJzn88MOZMGECJ510EpWVldxxxx3cdtttTJw4kQULFnDjjTdy6623ArB06VKOPvpoxo8fzznnnMPWrVtb9vm9732PSZMmccABB7BgwQIA3n33XSZNmsTEiRMZP348K1euTOfHljKpafWhErVIj930j3dZvm5HUvc5bngJP/jCwXGt29TUxBNPPMGpp54KwOLFi1m2bBljxozhzjvvpLS0lDfffJP6+nomT57MKaecwpIlS3jvvfdYvnw569evZ9y4cXzta1/bbb8bN27ksssuY/78+YwZM4YtW7YwYMAArrjiCoqKirjyyisBePbZZ1u2ufjii7n99ts54YQTuOGGG7jpppv4+c9/3hLnG2+8weOPP85NN93EM888wx133MHs2bO56KKLaGhoIBKJJOHTy7wUJeqGlOxWRFJn586dTJw4EfAl6q9//eu88sorTJo0qaUp2dNPP83bb7/dUv+8fft2Vq5cyfz587ngggsIhUIMHz6cE088cY/9v/baa0yZMqVlXwMGDOg0nu3bt7Nt2zZOOOEEAC655BLOP//8luXnnnsuAEcccQSVlZUAHHPMMdx8882sWbOGc889l/3337/nH0iApChR96wuS0SIu+SbbM111G0VFha2vHfOcfvttzN9+vTd1nn88cdTHd4ecnNzAX8TtLn+/MILL+Soo47in//8J6effjq//e1v271o9DYpqqNWiVqkL5o+fTq/+c1vaGz01Zvvv/8+NTU1TJkyhQceeIBIJMKnn37K888/v8e2Rx99NPPnz+fjjz8GYMuWLQAUFxdTVbVnS7HS0lL69+/fUv/8xz/+saV03ZGPPvqIfffdl+985zucddZZvP322wn9vkGRmhJ1NAINtZBTkJLdi0hmXHrppVRWVnL44YfjnKOsrIxHHnmEc845h+eee45x48YxatQojjnmmD22LSsr48477+Tcc88lGo0yePBg5s2bxxe+8AXOO+88Hn30UW6//fbdtvnDH/7AFVdcQW1tLfvuuy9z5szpNL4HH3yQP/7xj4TDYYYOHcq1116b1N8/U8w51/VKZv2Au4BDAAd8zTn3akfrVwwPuYXLVsKAfZMVp0iftmLFCg466KBMhyFp0t7f28wWOecq2ls/3qqPXwBPOucOBCYAK7rcYsence5aREQ602XVh5mVAlOAWQDOuQag60roKiVqEZFkiKdEPQbYCMwxsyVmdpeZFbZdycwuN7OFZuZbtitRi4gkRTyJOhs4HPiNc+4woAb4ftuVnHN3OucqnHMVWBZUfZbkUEVE9k7xJOo1wBrn3Ouxn/+GT9wdC4Vhx7oEQxMREYgjUTvnPgNWm9nY2KyTgI6HwgLICqtELSKSJPG2o/42cJ+Z5QAfAV/tdO1QDlSpRC0ikgxxNc9zzi2N1T+Pd86d7Zzb2ukGoWxfoo6jjbaI7J3uuecevvWtb/V423XrdhUGuxrzOlWWLl2alu7zqelCHsqBpjrY2Xk+F5Hgcc4Rjfb8KU3pGLGubaLuaszrRHQ2Dne6EnVqupBnxZ6uW/UZFHQ+QpaItPHE9+Gzd5K7z6GHwmk/6XBxZWUl06dP56ijjmLRokV88YtfZO7cudTX13POOedw0003UVlZyamnnsoRRxzB4sWLOfjgg7n33nspKCigvLycmTNnMm/ePK6++mqcc/zoRz/COccZZ5zBLbfcAsCcOXP48Y9/TL9+/ZgwYULLwEqzZs1ixowZnHfeeQAUFRVRXV0NwC233MKf/vQnsrKyOO2006ioqGDhwoVcdNFF5Ofn8+qrr3Laaadx6623UlFRwf3339/usYuKipg9ezZz584lPz+fRx99lCFDhrT7ecyaNYu8vDyWLFnC5MmT+dKXvsTs2bOpq6sjPz+fOXPmMGbMGG644QZ27tzJSy+9xDXXXMOMGTP49re/zbJly2hsbOTGG2/krLPOSvjPl6ISdXOiVj21SG+xcuVKvvnNb3Lbbbexdu1a3njjDZYuXcqiRYuYP38+AO+99x7f/OY3WbFiBSUlJfzv//5vy/YDBw5k8eLFTJkyhe9973s899xzLF26lDfffJNHHnmETz/9lB/84Ae8/PLLvPTSS3FVVTzxxBM8+uijvP7667z11ltcffXVnHfeeVRUVHDfffexdOlS8vPzW9Zft25du8cGqKmp4eijj+att95iypQp/O53v+v02GvWrOGVV17hZz/7GQceeCALFixgyZIl/PCHP+Taa68lJyeHH/7wh8ycOZOlS5cyc+ZMbr75Zk488UTeeOMNnn/+ea666ipqamp68NfYXWpK1KFWJWoR6Z5OSr6pNHr0aI4++miuvPJKnn76aQ477DAAqqurWblyJaNGjWLkyJFMnjwZgC9/+cv88pe/bBnwf+bMmQC8+eabTJ06lbKyMgAuuuiilkTfev7MmTN5//33O43pmWee4atf/SoFBX6At67GsO7o2GeffTY5OTnMmDED8GNYz5s3r9N9nX/++YRCIcCPjX3JJZewcuVKzKxl9MC2nn76aR577LGWJ9TU1dWxatWqhMdxSW3Vh8b7EOk1msedds5xzTXX8I1vfGO35ZWVlXs8Qbv1z63Hre6u7OzslnrxaDRKQ0Pyh0oOh8Mt8bYew7ojrX+f66+/nmnTpvHwww9TWVnJ1KlT293GOcdDDz3E2LFj213eU6mp+jCD/AHqRi7SC02fPp277767pY547dq1Lc9AXLVqFa++6gfO/POf/8xxxx23x/aTJk3ixRdfZNOmTUQiEe6//35OOOEEjjrqKF588UU2b95MY2Mjf/3rX1u2KS8vZ9GiRQA89thjLSXWz3/+88yZM6fliehdjWHd0bETtX37dvbZZx/A38hs1jaO6dOnc/vtt9M8KumSJUsSPjakKlEDFA9T1YdIL3TKKadw4YUXcswxx3DooYdy3nnntSSjsWPH8utf/5qDDjqIrVu38u///u97bD9s2DB+8pOfMG3aNCZMmMARRxzBWWedxbBhw7jxxhs55phjmDx58m7VAZdddhkvvvgiEyZM4NVXX20pzZ566qmceeaZVFRUMHHixJYqhVmzZnHFFVcwceJEdu7c2eWxE3X11VdzzTXXcNhhh+1WEp82bRrLly9n4sSJPPDAA1x//fU0NjYyfvx4Dj74YK6//vqEjw1xjkfdXRUVFW7hd0dD7Wa4/IWk71+kr+kN41FXVlYyY8YMli1blulQer1UjUfdfcVDVaIWEUmC1NxMBCgeDtXr/YNuQ6k7jIikR3l5eZ8rTd9888271ZWDb+1x3XXXZSii9qUwUQ8FF4WajVAyLGWHERHpqeuuuy5wSbk9qav6KBnup+r0IhKXVNwvkuDpyd85tXXUoHpqkTjk5eWxefNmJes+zjnH5s2bycvL69Z2qa2jBj1AQCQOI0aMYM2aNWzcuDHToUiK5eXlMWLEiG5tk7pEXTgILKQStUgcwuEwY8aMyXQYElCpq/rICsWa6Kl3oohIIlKXqEGJWkQkCVKcqIdpYCYRkQSlPlGrRC0ikpDUJuqSYVC3DRp3drmqiIi0L/UlalCpWkQkAam/mQhqoicikoAUJ2p1ehERSZRK1CIiAZfaRJ1XCuEC1VGLiCQgri7kZlYJVAERoKmjpxC0s6E6vYiIJKg7Y31Mc85t6vYRioer04uISAJSW/UBKlGLiCQo3kTtgKfNbJGZXd7eCmZ2uZktNLOFuw3VWBLrnahxdkVEeiTeRH2cc+5w4DTgP8xsStsVnHN3OucqnHMVZWVluxYUD4OmOt9DUUREui2uRO2cWxubbgAeBibFfYTm3omqpxYR6ZEuE7WZFZpZcfN74BQg/kcRqxu5iEhC4mn1MQR42Mya1/+zc+7JuI9QokQtIpKILhO1c+4jYEKPj1DU3DtRiVpEpCdS3zwvnAf5A9SNXESkh1KfqEFPehERSUCaErU6vYiI9FR6EnWJHsklItJT6av6qF4P0UhaDici0pekL1G7KFRvSMvhRET6kvQlalD1h4hID6SvjhqUqEVEekAlahGRgEtPoi4sAwupLbWISA+kJ1FnhaBoiHonioj0QHoSNcTaUq9L2+FERPqK9CXq4mEqUYuI9EB6E/UOlahFRLorjYl6qH8cV+POtB1SRKQvSGMd9XA/VfWHiEi3pLdEDWpLLSLSTemtowYlahGRbkp/olanFxGRbklfos4rhex8lahFRLopfYnaTA8QEBHpgfQlalCnFxGRHkh/olanFxGRbklzoh7qS9TOpfWwIiK9WXoTdclwaNrpeyiKiEhc4k7UZhYysyVmNrfHR2vp9KJ6ahGReHWnRD0bWJHQ0Ypj3chVTy0iEre4ErWZjQDOAO5K6GgqUYuIdFu8JeqfA1cD0YSO1tKNXCVqEZF4dZmozWwGsME5t6iL9S43s4VmtnDjxo3trxTOg/z+KlGLiHRDPCXqycCZZlYJ/AU40cz+1HYl59ydzrkK51xFWVlZx3srHq5ELSLSDV0maufcNc65Ec65cuBLwHPOuS/3+IjFQ3UzUUSkG9Lbjhpi432oRC0iEq/s7qzsnHsBeCGhIxYPg+r1EI1AViihXYmI7A3SX6IuHgouAjUd3HAUEZHdZCBRNzfRU/WHiEg80p+os/P8tKk+7YcWEemN0p+oQ2E/jTam/dAiIr1R+hN1Vuz+ZbQp7YcWEemNMpCoYyXqiBK1iEg8MpCoY03yVKIWEYmL6qhFRAJOddQiIgGnOmoRkYBTHbWISMBlro66ah3cfRp88GzaQxAR6U0yV0e97GFY9Qr8Y3baQxAR6U0yV0e9/h0/ze+f9hBERHqTzNVRN2usTXsIIiK9SebqqJs1KFGLiHQmc3XUAOFClahFRLqQuTpqgH4jlahFRLqQgUTd6pD9RkGkQZ1fREQ6kf5E3Vq/UX7aWJPRMEREgiwYiVo3FEVEOpTZRF00xE9VTy0i0qHMJuqcQj9tqM5oGCIiQZbZRJ1X6qd1OzIahohIkAUkUW/PaBgiIkHWZaI2szwze8PM3jKzd83spqQcOa8Uckv8eyVqEZEOZXe9CvXAic65ajMLAy+Z2RPOudd6fNT/fNfXTzvnf1aiFhHpUJeJ2jnngOa7feHYyyV01NIRftrc0aVeddQiIh2Jq47azEJmthTYAMxzzr2elKOHsiGnWCVqEZFOxJWonXMR59xEYAQwycwOabuOmV1uZgvNbOHGjRvjjyCvVIlaRKQT3Wr14ZzbBjwPnNrOsjudcxXOuYqysrL4d5pXokQtItKJeFp9lJlZv9j7fODzwL+SFoFK1CIinYqn1ccw4A9mFsIn9gedc3OTFkFeKexYl7TdiYj0NfG0+ngbOCxlEeSVwoYVKdu9iEhvl9meiaCqDxGRLmQ+UeeW+HbU0WimIxERCaTMJ+rCQeCiULct05GIiARS5hN18VA/1Q1FEZF2BSBRD/PTqs8yG4eISEAFKFF/mtk4REQCKgCJOlb1oUQtItKuzCfq7FwoGKhELSLSgcwnavDVHzuUqEVE2hOQRD1UJWoRkQ4EI1EXDYGabgyNKiKyFwlGos4t0ZPIRUQ6EIxEnVcCDVUQjWQ6EhGRwAlGom5+Gnl9VWbjEBEJoGAk6rzmRK3qDxGRtoKRqJtL1KqnFhHZQzASdV6pn6pELSKyh4AkapWoRUQ6EoxEndtFiXrNQlh0T9rCEREJkmAk6pYSdQeP5LrrJPjHbP8UmIf/HVa/ufvySJOeECMifVYwEnVuF4m6Wd02eOvP8IcZ/ucd6+DGUvivgfDar1MaoohIpgQjUYfzIJTT9c3ESIOfNtX56YfP7Vr2/lOpiU1EJMOCkaghvm7kbTvE1Gzy0+x8+OxtVX+ISJ8UnESdV9J1ibp11UjjTtj2CeT3h9P/xy/b8lFqYxQRyYAAJerS+Oqom/39clh4N5QdBCMq/LwP5qUsPBGRTAlOoi4YtKsqoy0L+WnrRL7iMT/d/2QoOxBGHgUL56Q2RhGRDOgyUZvZSDN73syWm9m7ZjY7JZEUDYbqDe0vy87z0/bqsIccAmbwuZNh0/vqNCMifU48Jeom4P8458YBRwP/YWbjkh5J0WD/8ADn9lyWneun7VWN9B/jp8MmAg7WL0t6aCIimdRlonbOfeqcWxx7XwWsAPZJeiSFgyHaCDu37rmspUQdS9QX/AUGjfXv+43y08EH+unG95IemohIJmV3Z2UzKwcOA15vZ9nlwOUAo0aN6n4khWV+WrMJCga0ibJNibpgIHz1Cdj0nm+DDVAyAsIFsP5diDQCBqFu/XoiIoEU981EMysCHgK+65zboyLYOXenc67COVdRVlbW/Ug6G0GvbYk6FIbCgTD62F3rZGXBfifCojnwX4Pg7undj0FEJIDiStRmFsYn6fucc39PSSTNibp1E7xmzSXq5iQeyml/H1/4JfQb7d+vXdjxzUkRkV4knlYfBvweWOGc+1nKImlJ1O3cMGwuUW9b5acdJerCgfAfb8CUq/zP/5qb3BhFRDIgnhL1ZOArwIlmtjT2Oj3pkbQk6vaqPmIl6o3/8tNQuOP9hLJh2nVQOhI+XpDcGEVEMqDLu23OuZcAS3kknZaoc3f/uaMSdTMzGDzOt6sWEenlgtMzMZwPWeGuu5FD14kaoOwA2LQSopHEYxMRyaDgJGoz3yyvtp1u5K7NqHidVX00KzsQIvWwtTIp4YmIZEpwEjVA8VCo+mzP+W0TdVYcibq5Q4w6wIhILxewRD0Mqj7dc37rbuVHXuarSbpSdoCfblyRnNhERDIkgIm6ixL1Gbf6apKu5JX6lh/r301efCIiGRC8RF2zEZoa/HjTK5/x89tWfcRryCGwbkn7Az2JiPQSAUvUQ/10ayW8/QDc92/+5+ZEO3py9/Z34On+qS+rXk1aiCIi6RawRD3MTz9dGpsRq+JwUSg/Hr76ePf2d9CZflr5cjKiExHJiGAl6pLmRP2Wn4YLYgtcfPXSbeX3860/Vr+WjOhERDIiYIk6Nsz1mjf9NCeWqF0UrIeh7nei70peX514fCIiGRCsRF0wwD87cXVsuOtwEhL1gaf7ji8fPpecGEVE0ixYiRp8j8JmzWN8JJKoRx3rm+qtfDrx2EREMiB4iXrfE3a9b66ucFF6PC5UKBtGHwcfz1czPRHplYKXqCu+BgP28+/rq/zUuZ6XqMFXf2z7BN57IvH4RETSLHiJunAQfGcxnPB9aKjyo98lUvUBMP5LUDQElt6XvDhFRNIkeIm6WW6xnzZUJ16iDmXDwefAynntP5hARCTAgpuo80r8tL4qVqJO8NkFh5znW3+s+EfisYmIpFFwE3VziTpZiXpEBQzYF97+S+KxiYikUfATdd0OfM/EBEM183XVHy+AbasTDk9EJF0CnKjbVn0kIdTxXwQcvPNg4vsSEUmTXpCodyQvUQ8YA6OOgSV/gsa6xPcnIpIGAU7Ubeqok/Ug9Mnf9UOfvnRbcvYnIpJivSBR70i8eV5rY0/1w5+++iuo3pCcfYqIpFBwE3VOkZ8ms4662Uk3QKQB/vl/1K1cRAIvuIk6KwtyimOJOoklaoBB+8O0a2HFY/Dqr5O3XxGRFMjuagUzuxuYAWxwzh2S+pBaySvxzfOSXaIGOPY7sGYhPH0d1G33iTvRttp7g2gENiyHVa/B2kVQvR4aavzF79AvQvlxkBXKdJQifUqXiRq4B/gVcG9qQ2lHbnGrVh9J3ndWCM7/A8ydDfP/G2o3wRk/S12y3rnVPxF943vQWAuNO/10nyPgwBmdH3fbalj5FGz52CfKkmFw6PlQMjw1sbZVvdF3FPrwOVj9ph+DBfz4KaUjIZwP7z7qW9MUDYGRk6DfaOg3yk8H7Av9YuuJSLd1maidc/PNrDwNsewpN1b1kYwOL+0JZcOZv4KCgfDyL6BwMEy7pmf72rkN1i6Egfv75GUGNZtg/TJ4/ylYeLfvwt6ahcBFYMihPrmF8/0Y3A01PgmHcuCzZbDsIWjaCdn5EAr7i9eL/wOjj4VoE+xY55N+VrZ/TboMJnxp1w3Z9tRshvf+CRv+5WPtNwoG7geDD/YPGTaDSCMsvheevcl/6yg7EMafDyOPhlFH+STcfIFp3OlHJ/zXXP8otZXzoKlNE8i8UigZAcVDoKneXyyz82Dfab4knp3b6p6B2/U+nAfFw/20re1rfCem4RN9fKn8VrRjHXzwDHz4vL+w1233r0gjlI6A/mOgf7lvBlp2oJ9Wb/BPLmp+WpFID8RTos6c3GL/j5CKqo9mZnDyTf4f6sWfwNaP4YTv+aQFULUePnoeNr3vS7YHn+PnL/ubr5Yp3Qcaan0yrt/ul2Vl+yTbWBs7RhaMn+lbmww/DHIK/dNrzGDpn2HRPfDuwz55Ne302zYnudwSOOTfYPJ3YNABfpsNK2DBz/w0FPYl1pwCX9resRYevxKeug4+dxLsf4qfluwDm1b6cbn/NRcqX/IXiXCBT4hNO3d9JvkDYMjBsH21fyJ8+fFw+v/A4IM6/hzD+XDIuf4Ffp81G2HrJ7DlQx/Xjk/9tOozf9ymBj8vnoc65JbA1GvgsIv8773oHv/5f/DMrgvg0PFw6o990k+Gqs/851S5wF8Mtnzo55eM8N8Qiof5Z3JmZfvPqvIlePsBoM0N6qxsH9PY0+GAU6H/6OTEJ3sNc3G0eoiVqOd2VkdtZpcDlwOMGjXqiE8++STx6B66zD+Wq7EWDvoCzEhh2+doxJccX/6lT5RHft1/jX/z97B9lS/9hgt2fe0vGOhLvdvX+GVjpsDEC/0/7LbVvlVJ6QgoG+urN/JKuxdP3XYfU15p9+p8nfP1x8sf9Ql5e6y7fCjHxwS+1D/uTBh3lk9u4C9Um1fC+uX+W8D6d32COe67PrmksqS6tRLWLm41pkvsWK1L60v+BJ+8DFlhiDZCfn//DWjEkX4M83WL4aWfw441virp8z/cdbHtjm2r4bXfwPtP7krMuSX+20v5cTDmBBh6aMefR1M9bFvlv1VsXw1FQ32d/vtPwab3/DplB/pnee53IoyerNK2AGBmi5xzFe0uS1aibq2iosItXLiwW0G266nrfKJsLq2d8dPE99mVbavhmR/A8sd8QsgtgdP+25dMw/m+esNCfpCn5keFBZVz/pvAB8/40uHgg2DkUT1LYEGwdhEs+zsUDYYjL/XfTFprqPWteF66zV+Uxp7mq3Ryivy6OYX+uZwDPwd5/XyyDRf46pd1i31peOmf/b4+d7JPzKOPhaETfDVZojZ/6C8AHzzrLzpNdf6CM/Uaf7EJhRPbfzTqq8Uijb5ws2GFvzhMnp147JJyvTdRv/wLmHeDL0UdMQvOuDXxfcarqd7/s2fnJf4PJOlV9ZlP1iv+4e8dNNbEt10oBw6/xCe2fiNTGiKNO+GTV/w5/vGLUFjmLw5Fg/1FpGysv8h8tgxWv+a/ue3c5qvRQuFY9VrYx5yVDVWf+m9DbX5Xh/H4qQtwBQMZVJTLgMIcmiKOuqYIedkhRg8soDDXX4QiUUck6sjJDm6r3b6ss0QdT/O8+4GpwCAzWwP8wDn3++SG2IHCwX4abUxdHXVHsnODX2KW9hUPhdNu8S/wJc3GWn+TtmYDbP7AP4/TxebXV8OQcb4E3d0qqp4K5/t7B/ud6EvYi/8AH70AtZt3VVE1y+vnb0zm9/cxRyO+NF5f5f83Io1QMAgO/4q/kZ2dC6EcHllTyLWvGrWPrAJWdRhKfjhEYyRKU9QX2gYV5ZAXDlHXGKG2IYJzMLgkl7Ii//9Q0xAhEo1SVpzL4OI8DKhpaKIx4hhQmMPg4lyys4yahggNTVE/rySXnFAWtQ0R6hoj9C/Ioawkl/xwiNqGJnY2RCnNDzO4JJfC3Gxq65uobYhQkh9mcHEuBTkh6pui1DZEKMwNkZu9dzUBjafVxwXpCKRdRYN3vU93opa+IysLcov8q3iIr2MOCjPY/2T/alZf7e8RbF/t67MHj/O/Qzcdu38dD1bUU5ibTWMkysaqerbUNBAOGbnhEDX1TXyyuZYtNQ3kZmeRk51Flhlrt+6kMRIlNxyiMCeEAzZW1bOxqp6sLOhXkEOWwcbqet74eAsARbnZZIeM5et2sLG6nkjUUZgTIjuUxfadjQl/TFkG0VZf/kvzw5Tmh9nZGKG2von8nBBlxXn0Lwi3XGBys7MoK85jYGEO9U0Rqusj5GQbZUW5DCzKpTESpbq+iZAZZcW5DCrKJRJ1VNc3YQZlxf7iFHVQU99ExLnYxSl2waqP0BSNMqgol4GFOYSyjJ2NERqbHMV52WRlJe++TrBbfShRy94ot8g3f+SohHYzuCSPwSW7mjQeMKST5ppJFI1l1OZE1RiJsqm6nqaIozA3m5zsLLbWNLChqo66xiiFudnkZvuEvqGqntr6Jgpzs8kPh1rm1dQ3UZAbIj8corquiQ1V9eyoa6QgJ5uCHF8q37Cjnm07/byBRbnUNUZYs7WWd9ZuIy8coiAnm4amCAuqNlFV10SWQWFONk1Rx87GSEK/c5b5tj4tLUpDxqCiXMKhLGobmqhvjNI/9m0jLxyiur6p5ZtF87eNzgQ7URe2TtTqNSjSG7QtSYZDWQwr3b2zU1FuNiMHZK61S2MkSnaWYWY450vRm6sbCGdnUZgTIupgQ1Udm6oaCGUZhbkhDGNjdR0bq+oxMwpzsgllwcbqBjbu8M1pC3Kzyc4yNlX7C1E06ijIzSYnlMWW2MWptqGJ4rxsBhXlsq22gSWrttHQFO003oAn6kG+JJ2MR3GJiMSEW5VgzYzivDDFebs3GhhQmAND226ZunsYdl3Hy4Jdn5AV8u2VQVUfIrLXCn72K9nHT5WoRWQvFfzs17/cT5WoRWQvFfzs15yoNcC/iOylek+irvoso2GIiGRK70nU2zruWSUi0pf1gkQdGxJSiVpE9lLBT9SlscFxjvxaZuMQEcmQYHd4AT9C2I3bMx2FiEjGBL9ELSKyl1OiFhEJOCVqEZGAU6IWEQk4JWoRkYBTohYRCTglahGRgFOiFhEJOHMpGJXOzKqA97q5WSnQnZ4t3V0fYBCwqZvb9OQ4ii24f8/uHqsncUH3Y+tLf8uebrO3x3awcy6/3SXOuaS/gIU92ObOVK6frrgUW7D/nt09Vk/i6klsfelvqdh6dhxgY0fLglT18Y8Ur99TPTmOYgvu37O7xwrq59XTbXpCsfVMd4+zraMFqar6WOicq0j6jhMU1LhAsfVEUOOC4MYW1LhAsXV2jFSVqO9M0X4TFdS4QLH1RFDjguDGFtS4QLF1eIyUlKhFRCR5glRHLSIi7VCiFhEJuB4najOrTmYgyWRmZ5uZM7MDMx1Le7r67MzsBTNL200VMxthZo+a2Uoz+9DMfmFmOZ2s/10zK0hjfIE814J+noHOtR7EF8hzra+WqC8AXopN42ZmodSEE1xmZsDfgUecc/sDBwBFwM2dbPZdIG3/PAGm86wbdK71XEKJ2syKzOxZM1tsZu+Y2Vmx+eVmtsLMfmdm75rZ02bWfo+bJDOzIuA44OvAl2LzpprZfDP7p5m9Z2Z3mFlWbFm1mf3UzN4CjklHjK1imtvq51+Z2ax0Hb+VE4E659wcAOdcBPhP4GtmVmhmt5rZMjN728y+bWbfAYYDz5vZ8+kKMmjnWm85z1rFpXMtTkE71yDxEnUdcI5z7nBgGvDT2FUTYH/g1865g/ENuf8twWPF6yzgSefc+8BmMzsiNn8S8G1gHLAfcG5sfiHwunNugnPupTTFGCQHA4taz3DO7QBWAZcC5cBE59x44D7n3C+BdcA059y0NMYZtHNN51n36VzroUQTtQE/MrO3gWeAfYAhsWUfO+eWxt4vwv8R0uEC4C+x939h19fSN5xzH8Wu4vfjS0MAEeChNMXW20wFfuucawJwzm3JYCxBO9d0niXXVHSudSjRp5BfBJQBRzjnGs2sEsiLLatvtV4ESMfX0QH4r1eHmpkDQoAD/hmbttb8c13snyrdmtj9QpnX0Yopthw4r/UMMysBRgGVmQioA4E513rZeQY617orMOdas0RL1KXAhtgvMw0YnYSYEnEe8Efn3GjnXLlzbiTwMXA8MMnMxsTqDGfibwJl0ifAODPLNbN+wEkZiuNZoMDMLoaWG10/Be4BngK+YWbZsWUDYttUAcVpjjNI51pvOs9A51p3BelcA3qYqGMfZj1wH1BhZu8AFwP/SmJsPXEB8HCbeQ/F5r8J/ApYgf+narteWjR/ds651cCDwLLYdEkm4nG+a+o5wPlmthJ4H19Hdy1wF77+8O3YTbALY5vdCTyZjhs8AT3XAn+egc617grouQb0sAu5mU0Afuecm5T8kJLPzKYCVzrnZmQ4lF732WVab/q8gnSeQe/67IIgyJ9Xt0vUZnYF/ibJ/01+OH2bPrvu0efVc/rsuifon5cGZRIRCbi4StRmNtLMnjez5bGG3rNj8weY2Tzz3UHnmVn/2PyLYo3W3zGzV2JfKZr3dWqsM8AHZvb91Pxa0lsl+Vy728w2mNmyTP0+EkzJOs862k/SxfmImGHA4bH3xfibAOOA/wa+H5v/feCW2Ptjgf6x96fhG/qDb8b0IbAvkAO8BYzr7iNu9Oq7r2Sda7GfpwCHA8sy/XvpFaxXEnNau/tJerw9/CUfBT6Pf4DtsFYBv9fOuv2BtbH3xwBPtVp2DXBNpv9oegX31dNzrdW8ciVqvbp6JXqetd1PsuPryc3EcuAw4HVgiHPu09iiz9jVe6e1rwNPxN7vA6xutWxNbJ7IHhI810TikqzzrM1+kqpbPRPND0TzEPBd59yOXd3ffRvJWC+t1utPw/9SxyHSDTrXJB2SdZ613U+y44y7RG1m4Vgg9znn/h6bvd7MhsWWDwM2tFp/PL4R+1nOuc2x2WuBka12OyI2T6RFks41kU4l6zzrYD9JFW+rDwN+D6xwzv2s1aLHgEti7y/B189gZqPw485+xfnRxZq9Cewf62Kbgx8e8rHEfgXpS5J4rol0KFnnWSf7SW68sQrwzlcyOw5YALwDRGOzr8XXxTyIH1TlE+CLzrktZnYXfvi/T2LrNrnYY9DN7HTg5/gWIHc75zobNFz2Mkk+1+7Hj8o2CFgP/MA59/s0/SoSYMk6zzraj3Pu8aTGG0+iFhGRzOmrj+ISEekzlKhFRAJOiVpEJOCUqEVEAk6JWkQk4JSopc8xsxvN7MpOlp9tZuPSGZNIIpSoZW90Nn6kNJFeQe2opU8ws+vwPck24Af+WgRsBy7HD6n7AfAVYCIwN7ZsO74TA8Cv8U+ergUuc85l/Dl5Is2UqKXXM7Mj8E+yPgo/0Nhi4A5gTvOYDGb2/4D1zrnbzeweYK5z7m+xZc8CVzjnVprZUcCPnXMnpv83EWlft0bPEwmo44GHnXO1AGbWPH7MIbEE3Q8oAp5qu2Fs1LNjgb+2GjktN9UBi3SHErX0ZfcAZzvn3jKzWfhxP9rKArY55yamLyyR7tHNROkL5gNnm1m+mRUDX4jNLwY+jQ1DeVGr9atiy4iNHfyxmZ0PfjS01s9dFAkCJWrp9Zxzi4EH8M/gfAI/nC7A9fjR0F4GWt8c/AtwlZktMbP98En862b2FvAucFa6YheJh24miogEnErUIiIBp0QtIhJwStQiIgGnRC0iEnBK1CIiAadELSIScErUIiIBp0QtIhJw/x99EWsxZRZ6BgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "start = len(rrate_df)-1\n",
    "end = len(rrate_df)+len(future_projections) - 1\n",
    "  \n",
    "# Predictions against the test set\n",
    "rrate_predictions = result.predict(start, end,\n",
    "                             typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "# plot predictions and actual values\n",
    "rrate_predictions.plot(legend = True)\n",
    "rrate_df['reproduction_rate'].plot(legend = True)\n",
    "\n",
    "result.predict(start, end,typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "future_projections['reproduction_rate'] = rrate_predictions\n",
    "future_projections"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecd8f290",
   "metadata": {},
   "source": [
    "We see that the trend is a declining which is on par with our data so far"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea463df5",
   "metadata": {},
   "source": [
    "###  Synthetically projecting positive_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "37403569",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "fd7d4544",
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "cannot insert positive_rate, already exists",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-35-90d5b5efbb7c>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mfuture_projections\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0minsert\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mloc\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m4\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mcolumn\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m'positive_rate'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mvalue\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0.0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfuture_projections\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mpr_df\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0msample_df\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdrop\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msample_df\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdifference\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'positive_rate'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minplace\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\users\\bhara\\anaconda3\\envs\\aimed\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36minsert\u001b[1;34m(self, loc, column, value, allow_duplicates)\u001b[0m\n\u001b[0;32m   3761\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_ensure_valid_index\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mvalue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3762\u001b[0m         \u001b[0mvalue\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_sanitize_column\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcolumn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mbroadcast\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3763\u001b[1;33m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_mgr\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0minsert\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mloc\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcolumn\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mallow_duplicates\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mallow_duplicates\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3764\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3765\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0massign\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m->\u001b[0m \u001b[0mDataFrame\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\users\\bhara\\anaconda3\\envs\\aimed\\lib\\site-packages\\pandas\\core\\internals\\managers.py\u001b[0m in \u001b[0;36minsert\u001b[1;34m(self, loc, item, value, allow_duplicates)\u001b[0m\n\u001b[0;32m   1189\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mallow_duplicates\u001b[0m \u001b[1;32mand\u001b[0m \u001b[0mitem\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1190\u001b[0m             \u001b[1;31m# Should this be a different kind of error??\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1191\u001b[1;33m             \u001b[1;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf\"cannot insert {item}, already exists\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1192\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1193\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mloc\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mint\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mValueError\u001b[0m: cannot insert positive_rate, already exists"
     ]
    }
   ],
   "source": [
    "future_projections.insert(loc=4,column='positive_rate',value=[0.0]*len(future_projections))\n",
    "pr_df = sample_df.drop(sample_df.columns.difference(['positive_rate']), 1, inplace=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "f205c51a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing stepwise search to minimize aic\n",
      " ARIMA(1,1,1)(0,1,1)[12]             : AIC=-4327.317, Time=1.16 sec\n",
      " ARIMA(0,1,0)(0,1,0)[12]             : AIC=-4177.824, Time=0.12 sec\n",
      " ARIMA(1,1,0)(1,1,0)[12]             : AIC=-4283.873, Time=0.88 sec\n",
      " ARIMA(0,1,1)(0,1,1)[12]             : AIC=-4325.747, Time=0.96 sec\n",
      " ARIMA(1,1,1)(0,1,0)[12]             : AIC=-4193.613, Time=0.28 sec\n",
      " ARIMA(1,1,1)(1,1,1)[12]             : AIC=-4317.333, Time=1.93 sec\n",
      " ARIMA(1,1,1)(0,1,2)[12]             : AIC=-4337.073, Time=4.03 sec\n",
      " ARIMA(1,1,1)(1,1,2)[12]             : AIC=-4336.527, Time=2.75 sec\n",
      " ARIMA(0,1,1)(0,1,2)[12]             : AIC=-4335.431, Time=3.28 sec\n",
      " ARIMA(1,1,0)(0,1,2)[12]             : AIC=-4329.051, Time=3.16 sec\n",
      " ARIMA(2,1,1)(0,1,2)[12]             : AIC=-4353.937, Time=5.18 sec\n",
      " ARIMA(2,1,1)(0,1,1)[12]             : AIC=-4344.738, Time=1.78 sec\n",
      " ARIMA(2,1,1)(1,1,2)[12]             : AIC=-4366.966, Time=9.88 sec\n",
      " ARIMA(2,1,1)(1,1,1)[12]             : AIC=-4335.089, Time=1.70 sec\n",
      " ARIMA(2,1,1)(2,1,2)[12]             : AIC=-4335.760, Time=3.66 sec\n",
      " ARIMA(2,1,1)(2,1,1)[12]             : AIC=-4328.779, Time=3.55 sec\n",
      " ARIMA(2,1,0)(1,1,2)[12]             : AIC=-4354.890, Time=1.35 sec\n",
      " ARIMA(3,1,1)(1,1,2)[12]             : AIC=-4351.618, Time=2.01 sec\n",
      " ARIMA(2,1,2)(1,1,2)[12]             : AIC=-4353.267, Time=5.34 sec\n",
      " ARIMA(1,1,0)(1,1,2)[12]             : AIC=-4328.485, Time=2.82 sec\n",
      " ARIMA(1,1,2)(1,1,2)[12]             : AIC=-4345.999, Time=5.15 sec\n",
      " ARIMA(3,1,0)(1,1,2)[12]             : AIC=-4353.545, Time=5.28 sec\n",
      " ARIMA(3,1,2)(1,1,2)[12]             : AIC=-4349.720, Time=9.10 sec\n",
      " ARIMA(2,1,1)(1,1,2)[12] intercept   : AIC=-4351.799, Time=8.35 sec\n",
      "\n",
      "Best model:  ARIMA(2,1,1)(1,1,2)[12]          \n",
      "Total fit time: 83.712 seconds\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>                    <td>y</td>                  <th>  No. Observations:  </th>    <td>539</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>SARIMAX(2, 1, 1)x(1, 1, [1, 2], 12)</td> <th>  Log Likelihood     </th> <td>2190.483</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                     <td>Sun, 27 Jun 2021</td>           <th>  AIC                </th> <td>-4366.966</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                         <td>13:49:07</td>               <th>  BIC                </th> <td>-4337.109</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                           <td>0</td>                  <th>  HQIC               </th> <td>-4355.276</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                               <td> - 539</td>                <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>                 <td>opg</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>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1</th>    <td>   -0.2564</td> <td>    0.120</td> <td>   -2.134</td> <td> 0.033</td> <td>   -0.492</td> <td>   -0.021</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L2</th>    <td>   -0.2175</td> <td>    0.028</td> <td>   -7.889</td> <td> 0.000</td> <td>   -0.271</td> <td>   -0.163</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>    <td>    0.0264</td> <td>    0.127</td> <td>    0.208</td> <td> 0.835</td> <td>   -0.222</td> <td>    0.275</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.S.L12</th> <td>   -0.2783</td> <td>    0.503</td> <td>   -0.554</td> <td> 0.580</td> <td>   -1.263</td> <td>    0.707</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L12</th> <td>   -0.3358</td> <td>    0.502</td> <td>   -0.669</td> <td> 0.503</td> <td>   -1.319</td> <td>    0.647</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L24</th> <td>   -0.2225</td> <td>    0.317</td> <td>   -0.702</td> <td> 0.483</td> <td>   -0.844</td> <td>    0.399</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>   <td> 1.395e-05</td> <td> 3.95e-07</td> <td>   35.343</td> <td> 0.000</td> <td> 1.32e-05</td> <td> 1.47e-05</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.05</td> <th>  Jarque-Bera (JB):  </th> <td>3784.80</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.83</td> <th>  Prob(JB):          </th>  <td>0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>0.07</td> <th>  Skew:              </th>  <td>0.67</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.00</td> <th>  Kurtosis:          </th>  <td>16.07</td> \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                        SARIMAX Results                                        \n",
       "===============================================================================================\n",
       "Dep. Variable:                                       y   No. Observations:                  539\n",
       "Model:             SARIMAX(2, 1, 1)x(1, 1, [1, 2], 12)   Log Likelihood                2190.483\n",
       "Date:                                 Sun, 27 Jun 2021   AIC                          -4366.966\n",
       "Time:                                         13:49:07   BIC                          -4337.109\n",
       "Sample:                                              0   HQIC                         -4355.276\n",
       "                                                 - 539                                         \n",
       "Covariance Type:                                   opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ar.L1         -0.2564      0.120     -2.134      0.033      -0.492      -0.021\n",
       "ar.L2         -0.2175      0.028     -7.889      0.000      -0.271      -0.163\n",
       "ma.L1          0.0264      0.127      0.208      0.835      -0.222       0.275\n",
       "ar.S.L12      -0.2783      0.503     -0.554      0.580      -1.263       0.707\n",
       "ma.S.L12      -0.3358      0.502     -0.669      0.503      -1.319       0.647\n",
       "ma.S.L24      -0.2225      0.317     -0.702      0.483      -0.844       0.399\n",
       "sigma2      1.395e-05   3.95e-07     35.343      0.000    1.32e-05    1.47e-05\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   0.05   Jarque-Bera (JB):              3784.80\n",
       "Prob(Q):                              0.83   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               0.07   Skew:                             0.67\n",
       "Prob(H) (two-sided):                  0.00   Kurtosis:                        16.07\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from pmdarima import auto_arima\n",
    "  \n",
    "# Ignore harmless warnings\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "\n",
    "## We try to find the best arima model for our data\n",
    "# Fit auto_arima function to our dataset\n",
    "stepwise_fit = auto_arima(pr_df['positive_rate'], start_p = 1, start_q = 1,\n",
    "                          max_p = 5, max_q = 3, m = 12,\n",
    "                          start_P = 0, seasonal = True,\n",
    "                          d = None, D = 1, trace = True,\n",
    "                          error_action ='ignore',   # we don't want to know if an order does not work\n",
    "                          suppress_warnings = True,  # we don't want convergence warnings\n",
    "                          stepwise = True)           # set to stepwise\n",
    "  \n",
    "# To print the summary\n",
    "stepwise_fit.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "2fd53098",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='date'>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "split_date = '01-May-2021'\n",
    "\n",
    "pr_df_train = pr_df.loc[pr_df.index <= split_date].copy()\n",
    "pr_df_test = pr_df.loc[pr_df.index > split_date].copy()\n",
    "\n",
    "# Fit the best SARIMAX model obtained from above on the training set\n",
    "from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
    "\n",
    "model = SARIMAX(pr_df_train['positive_rate'], \n",
    "                order = (2, 1, 1), \n",
    "                seasonal_order =(1, 1, [1,2], 12))\n",
    "  \n",
    "result = model.fit()\n",
    "#result.summary()\n",
    "\n",
    "start = len(pr_df_train)\n",
    "end = len(pr_df_train) + len(pr_df_test) - 1\n",
    "  \n",
    "# Predictions against the test set\n",
    "predictions = result.predict(start, end,\n",
    "                             typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "# plot predictions and actual values\n",
    "predictions.plot(legend = True)\n",
    "pr_df_test['positive_rate'].plot(legend = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2daac575",
   "metadata": {},
   "source": [
    "The SARIMAX model has gathered an overall trend in our data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f7d1dccb",
   "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>days_from_start</th>\n",
       "      <th>vaccination_ratio</th>\n",
       "      <th>reproduction_rate</th>\n",
       "      <th>stringency_index</th>\n",
       "      <th>positive_rate</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</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>2021-06-22</th>\n",
       "      <td>172</td>\n",
       "      <td>0.256474</td>\n",
       "      <td>0.886064</td>\n",
       "      <td>51.868969</td>\n",
       "      <td>0.083573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-23</th>\n",
       "      <td>173</td>\n",
       "      <td>0.258015</td>\n",
       "      <td>0.887335</td>\n",
       "      <td>51.790031</td>\n",
       "      <td>0.082886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-24</th>\n",
       "      <td>174</td>\n",
       "      <td>0.259557</td>\n",
       "      <td>0.888973</td>\n",
       "      <td>51.771781</td>\n",
       "      <td>0.083282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-25</th>\n",
       "      <td>175</td>\n",
       "      <td>0.261098</td>\n",
       "      <td>0.889609</td>\n",
       "      <td>51.611657</td>\n",
       "      <td>0.083173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-06-26</th>\n",
       "      <td>176</td>\n",
       "      <td>0.262640</td>\n",
       "      <td>0.888440</td>\n",
       "      <td>51.577481</td>\n",
       "      <td>0.083199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-25</th>\n",
       "      <td>420</td>\n",
       "      <td>0.638754</td>\n",
       "      <td>0.806695</td>\n",
       "      <td>43.853864</td>\n",
       "      <td>0.034274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-26</th>\n",
       "      <td>421</td>\n",
       "      <td>0.640295</td>\n",
       "      <td>0.806359</td>\n",
       "      <td>43.899959</td>\n",
       "      <td>0.034428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-27</th>\n",
       "      <td>422</td>\n",
       "      <td>0.641837</td>\n",
       "      <td>0.806788</td>\n",
       "      <td>43.895360</td>\n",
       "      <td>0.034184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-02-28</th>\n",
       "      <td>423</td>\n",
       "      <td>0.643378</td>\n",
       "      <td>0.807659</td>\n",
       "      <td>43.979936</td>\n",
       "      <td>0.034013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-03-01</th>\n",
       "      <td>424</td>\n",
       "      <td>0.644920</td>\n",
       "      <td>0.808710</td>\n",
       "      <td>43.949724</td>\n",
       "      <td>0.034115</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>253 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            days_from_start  vaccination_ratio  reproduction_rate  \\\n",
       "date                                                                \n",
       "2021-06-22              172           0.256474           0.886064   \n",
       "2021-06-23              173           0.258015           0.887335   \n",
       "2021-06-24              174           0.259557           0.888973   \n",
       "2021-06-25              175           0.261098           0.889609   \n",
       "2021-06-26              176           0.262640           0.888440   \n",
       "...                     ...                ...                ...   \n",
       "2022-02-25              420           0.638754           0.806695   \n",
       "2022-02-26              421           0.640295           0.806359   \n",
       "2022-02-27              422           0.641837           0.806788   \n",
       "2022-02-28              423           0.643378           0.807659   \n",
       "2022-03-01              424           0.644920           0.808710   \n",
       "\n",
       "            stringency_index  positive_rate  \n",
       "date                                         \n",
       "2021-06-22         51.868969       0.083573  \n",
       "2021-06-23         51.790031       0.082886  \n",
       "2021-06-24         51.771781       0.083282  \n",
       "2021-06-25         51.611657       0.083173  \n",
       "2021-06-26         51.577481       0.083199  \n",
       "...                      ...            ...  \n",
       "2022-02-25         43.853864       0.034274  \n",
       "2022-02-26         43.899959       0.034428  \n",
       "2022-02-27         43.895360       0.034184  \n",
       "2022-02-28         43.979936       0.034013  \n",
       "2022-03-01         43.949724       0.034115  \n",
       "\n",
       "[253 rows x 5 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "start = len(rrate_df)-1\n",
    "end = len(rrate_df)+len(future_projections) - 1\n",
    "\n",
    "# Predictions against the test set\n",
    "pr_predictions = result.predict(start, end,\n",
    "                             typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "# plot predictions and actual values\n",
    "pr_predictions.plot(legend = True)\n",
    "pr_df['positive_rate'].plot(legend = True)\n",
    "\n",
    "result.predict(start, end,typ = 'levels').rename(\"Predictions\")\n",
    "\n",
    "future_projections['positive_rate'] = pr_predictions\n",
    "future_projections"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "468d73b7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "784b4a56",
   "metadata": {},
   "outputs": [],
   "source": [
    "future_projections = future_projections.drop(\"days_from_start\",axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "86471aeb",
   "metadata": {},
   "outputs": [],
   "source": [
    "future_projections.to_csv('Prepared_data/Projection_Global_PreProcessed_dataset_unscaled.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "80bbfb91",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "98cc3ac6",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}