Time_series_prediction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 796,
   "id": "4e5a2fbb",
   "metadata": {},
   "outputs": [
    {
     "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": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import date\n",
    "#from matplotlib.dates import DateFormatter\n",
    "#import matplotlib.dates as mdates\n",
    "import seaborn as sns\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.graphics.tsaplots import plot_acf, plot_pacf\n",
    "#import datetime\n",
    "from scipy import stats\n",
    "from scipy.signal import savgol_filter\n",
    "from scipy.optimize import curve_fit\n",
    "from statsmodels.tsa.stattools import grangercausalitytests\n",
    "from statsmodels.tsa.stattools import adfuller\n",
    "from statsmodels.tsa.stattools import coint\n",
    "from statsmodels.tsa.vector_ar.var_model import VAR\n",
    "#from statsmodels.tsa.arima.model import ARIMA\n",
    "from pmdarima.arima import auto_arima\n",
    "from statsmodels.tsa.vector_ar.vecm import VECM, select_coint_rank\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "def mape(actual, fc): \n",
    "    actual, fc = np.array(actual), np.array(fc)\n",
    "    return np.mean(np.abs((actual - fc) / actual)) * 100\n",
    "\n",
    "def func_exp(t, tau):\n",
    "    return np.exp(-t / tau)\n",
    "\n",
    "df1 = pd.read_csv(\"C:\\\\Users\\\\Ping\\\\Desktop\\\\Project\\\\COVID_prediction\\\\data\\\\gtrend_caution.csv\") \n",
    "df1.head()\n",
    "df1['date'] = pd.to_datetime(df1.date)\n",
    "df1.set_index(\"date\", inplace = True)\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(df1.caution,'k-')\n",
    "ax.set_xlabel(\"Time\")\n",
    "ax.set_ylabel(\"Relative Caution\")\n",
    "ax.set_title(\"Caution Index (New York)\")\n",
    "\n",
    "# Smoothing\n",
    "df1['caution_sm'] = savgol_filter(df1.caution, window_length = 11, polyorder = 2)\n",
    "ax.plot(df1.caution_sm,'r-')\n",
    "ax.legend(['Original','Smoothed'])\n",
    "plt.show()\n",
    "\n",
    "df1['caution'] = df1['caution_sm']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 797,
   "id": "30875767",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([18262., 18353., 18444., 18536., 18628., 18718., 18809., 18901.,\n",
       "        18993.]),\n",
       " [Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, ''),\n",
       "  Text(0, 0, '')])"
      ]
     },
     "execution_count": 797,
     "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": [
    "#Read in COVID case numbers\n",
    "\n",
    "df2 = pd.read_csv('C:\\\\Users\\\\Ping\\\\Desktop\\\\Project\\\\COVID_prediction\\\\data\\\\United_States_COVID-19_Cases_and_Deaths_by_State_over_Time.csv') \n",
    "df2['submission_date']=pd.to_datetime(df2.submission_date)\n",
    "df2.set_index('submission_date', inplace = True)\n",
    "df2 = df2[df2.state==\"NY\"]\n",
    "#df2 = df2[df2.state==\"CA\"]\n",
    "df2.head()\n",
    "\n",
    "# Aggregate to weekly\n",
    "new_cases = df2.new_case.resample('W').mean()\n",
    "\n",
    "plt.plot(new_cases,'k-')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('New cases per day')\n",
    "plt.title('COVID-19 cases (New York)')\n",
    "plt.xticks(rotation=45)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 798,
   "id": "87ea1c69",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pandas\\core\\arrays\\datetimes.py:741: PerformanceWarning: Non-vectorized DateOffset being applied to Series or DatetimeIndex\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>temp_C</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>month_date</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-01-20</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-27</th>\n",
       "      <td>1.639374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-03</th>\n",
       "      <td>1.431490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-10</th>\n",
       "      <td>1.450395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-17</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-11-20</th>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-11-27</th>\n",
       "      <td>9.317951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-04</th>\n",
       "      <td>8.822310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-11</th>\n",
       "      <td>8.498068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-18</th>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>205 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              temp_C\n",
       "month_date          \n",
       "2019-01-20  2.000000\n",
       "2019-01-27  1.639374\n",
       "2019-02-03  1.431490\n",
       "2019-02-10  1.450395\n",
       "2019-02-17  2.000000\n",
       "...              ...\n",
       "2022-11-20  9.000000\n",
       "2022-11-27  9.317951\n",
       "2022-12-04  8.822310\n",
       "2022-12-11  8.498068\n",
       "2022-12-18  9.000000\n",
       "\n",
       "[205 rows x 1 columns]"
      ]
     },
     "execution_count": 798,
     "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": [
    "# Repeat average temperature data over several years and interpolate to weeks\n",
    "\n",
    "df3 = pd.read_csv('C:\\\\Users\\\\Ping\\\\Desktop\\\\Project\\\\COVID_prediction\\\\data\\\\ny_weather.csv') \n",
    "df3['month_date']=pd.to_datetime(df3.month_date)\n",
    "df3.set_index('month_date', inplace = True)\n",
    "\n",
    "# Other way\n",
    "#df3['yyyy'] = pd.to_datetime(df3['month_date']).dt.year\n",
    "#df3['mm'] = pd.to_datetime(df3['month_date']).dt.month\n",
    "#df3['dd'] = pd.to_datetime(df3['month_date']).dt.day\n",
    "\n",
    "#duplicate to multiyear\n",
    "#df3_copy1 = df3.copy()\n",
    "#df3_copy1['yyyy'] = df3_copy1['yyyy']-1\n",
    "#df3_copy1['month_date'] = df3['mm'].astype(str) +'/'+ df3['dd'].astype(str) + '/'+ df3_copy1['yyyy'].astype(str)\n",
    "\n",
    "#df3_copy2 = df3.copy()\n",
    "#df3_copy2['yyyy'] = df3_copy2['yyyy']+1\n",
    "#df3_copy2['month_date'] = df3['mm'].astype(str) +'/'+ df3['dd'].astype(str) + '/'+ df3_copy2['yyyy'].astype(str)\n",
    "\n",
    "#df3_copy3 = df3.copy()\n",
    "#df3_copy3['yyyy'] = df3_copy3['yyyy']+2\n",
    "#df3_copy3['month_date'] = df3['mm'].astype(str) +'/'+ df3['dd'].astype(str) + '/'+ df3_copy3['yyyy'].astype(str)\n",
    "\n",
    "df3_copy1 = df3.copy()\n",
    "df3_copy1.index = df3_copy1.index + pd.DateOffset(year=2019)\n",
    "df3_copy2 = df3.copy()\n",
    "df3_copy2.index = df3_copy2.index + pd.DateOffset(year=2021)\n",
    "df3_copy3 = df3.copy()\n",
    "df3_copy3.index = df3_copy3.index + pd.DateOffset(year=2022)\n",
    "\n",
    "df3_long = pd.concat([df3_copy1,df3,df3_copy2,df3_copy3],axis=0)\n",
    "#df3_long['month_date']=pd.to_datetime(df3_long.month_date)\n",
    "#df3_long.set_index('month_date', inplace = True)\n",
    "\n",
    "df3_long = df3_long.resample('W').mean().interpolate(method='spline',order=3)\n",
    "\n",
    "plt.plot(df3_long)\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "df3_long"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 799,
   "id": "9a5129a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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>caution</th>\n",
       "      <th>caution_sm</th>\n",
       "      <th>new_case</th>\n",
       "      <th>temp_C</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [caution, caution_sm, new_case, temp_C]\n",
       "Index: []"
      ]
     },
     "execution_count": 799,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merge the full date range into previous\n",
    "df_merged = pd.merge(df1,new_cases, how='inner',left_index=True,right_index=True)\n",
    "df_merged = pd.merge(df_merged,df3_long, how='inner',left_index=True,right_index=True)\n",
    "\n",
    "# Coplot full range\n",
    "fig,ax = plt.subplots()\n",
    "ax.plot(new_cases,color='royalblue')\n",
    "ax.set_ylabel('New cases per day',color='royalblue')\n",
    "ax.set_xlabel('Time')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "ax2 = ax.twinx()\n",
    "ax2.plot(df_merged.caution,color='orangered')\n",
    "ax2.set_ylabel('Relative caution',color='orangered')\n",
    "\n",
    "plt.show()\n",
    "\n",
    "fig.savefig('Caution_and_new_cases.png',\n",
    "            format='png',\n",
    "            dpi=100,\n",
    "            bbox_inches='tight')\n",
    "\n",
    "# Subtract baseline \"caution\" from pre-pandemic and renormalize\n",
    "df_merged.caution = df_merged.caution-df_merged.caution['2020-01-26':'2020-02-16'].mean()\n",
    "df_merged.caution = (df_merged.caution-df_merged.caution.min())/(df_merged.caution.max()-df_merged.caution.min())\n",
    "\n",
    "# Skip initial period of pandemic\n",
    "df_during = df_merged['2020-04-10':]\n",
    "# Check for any embedded NaN's that need to be filled\n",
    "df_during[df_during.caution.isna()==True]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 800,
   "id": "352ee799",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:1667: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self.obj[key] = value\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Coplot \n",
    "fig,ax = plt.subplots()\n",
    "ax.plot(df_during.new_case,color='royalblue')\n",
    "ax.set_ylabel('New cases per day',color='royalblue')\n",
    "ax.set_xlabel('Time')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "ax2 = ax.twinx()\n",
    "ax2.plot(df_during.caution,color='orangered')\n",
    "ax2.set_ylabel('Relative caution',color='orangered')\n",
    "\n",
    "np.random.seed(0)\n",
    "t = np.arange(0,len(df_during.caution))\n",
    "params, cov = curve_fit(func_exp, t, df_during.caution.values)\n",
    "tau = params\n",
    "y_fit = func_exp(t, tau)\n",
    "\n",
    "df_during.loc[:,'caution2'] = df_during.caution-y_fit\n",
    "\n",
    "ax2.plot(df_during.index, y_fit, color='orangered',linestyle = '--')\n",
    "\n",
    "plt.show()\n",
    "\n",
    "#fig.savefig('Caution_and_new_cases.png',\n",
    "#            format='png',\n",
    "#            dpi=100,\n",
    "#            bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 801,
   "id": "8ea5e44a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/3332752693.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_during['caution2']= (df_during.caution2-min_cau)/(range_cau)\n",
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/3332752693.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_during['new_case']= (df_during.new_case-df_during.new_case.min())/(df_during.new_case.max()-df_during.new_case.min())\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/3332752693.py:21: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_during['log_caution'] = np.log(df_during['caution2']+epsilon)\n",
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/3332752693.py:22: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_during['log_new_case'] = np.log(df_during['new_case']+epsilon)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.79519292],\n",
       "       [0.79519292, 1.        ]])"
      ]
     },
     "execution_count": 801,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Max-min normalize\n",
    "min_cau = np.min(df_during.caution2)\n",
    "range_cau = np.max(df_during.caution2)-np.min(df_during.caution2)\n",
    "df_during['caution2']= (df_during.caution2-min_cau)/(range_cau)\n",
    "df_during['new_case']= (df_during.new_case-df_during.new_case.min())/(df_during.new_case.max()-df_during.new_case.min())\n",
    "\n",
    "fig,ax = plt.subplots()\n",
    "ax.plot(df_during.new_case,color='royalblue')\n",
    "ax.set_ylabel('New cases per day',color='royalblue')\n",
    "ax.set_xlabel('Time')\n",
    "plt.xticks(rotation=45)\n",
    "#ax.plot(df_during.new_case.diff(),color='black')\n",
    "\n",
    "ax2 = ax.twinx()\n",
    "ax2.plot(df_during.caution2,color='orangered')\n",
    "ax2.set_ylabel('Relative caution',color='orangered')\n",
    "\n",
    "plt.show()\n",
    "\n",
    "epsilon = 0.01  # Prevent -inf\n",
    "df_during['log_caution'] = np.log(df_during['caution2']+epsilon)\n",
    "df_during['log_new_case'] = np.log(df_during['new_case']+epsilon)\n",
    "\n",
    "np.corrcoef(df_during.log_caution, df_during.log_new_case)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 802,
   "id": "bbcd79a4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(-2.2160987620139925, 0.2004545044060378, 9, 76, {'1%': -3.5194805351545413, '5%': -2.9003945086747343, '10%': -2.5874984279778395}, -329.79882457868416)\n",
      "(-2.1408716722765755, 0.22840050660386335, 1, 84, {'1%': -3.510711795769895, '5%': -2.8966159448223734, '10%': -2.5854823866213152}, -224.2073270915622)\n",
      "(-3.436870064817821, 0.009764378072376909, 8, 76, {'1%': -3.5194805351545413, '5%': -2.9003945086747343, '10%': -2.5874984279778395}, -458.81192529594705)\n",
      "(-5.12617783308809, 1.2415723706919786e-05, 0, 84, {'1%': -3.510711795769895, '5%': -2.8966159448223734, '10%': -2.5854823866213152}, -217.29635760126746)\n",
      "(-2.7289189542360175, 0.06916631530496035, 2, 83, {'1%': -3.5117123057187376, '5%': -2.8970475206326833, '10%': -2.5857126912469153}, 21.10161040438203)\n",
      "(-3.076627327456718, 0.028331922005594215, 2, 83, {'1%': -3.5117123057187376, '5%': -2.8970475206326833, '10%': -2.5857126912469153}, -52.39264297525543)\n"
     ]
    }
   ],
   "source": [
    "print(sm.tsa.stattools.adfuller(df_during.caution2))\n",
    "print(sm.tsa.stattools.adfuller(df_during.new_case))\n",
    "\n",
    "# One round of differencing makes stationary\n",
    "print(sm.tsa.stattools.adfuller(df_during.caution.diff().dropna()))\n",
    "print(sm.tsa.stattools.adfuller(df_during.new_case.diff().dropna()))\n",
    "\n",
    "# Log transform also works\n",
    "print(sm.tsa.stattools.adfuller(df_during.log_caution))\n",
    "print(sm.tsa.stattools.adfuller(df_during.log_new_case))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 770,
   "id": "2f8993b0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 1\n",
      "ssr based F test:         F=8.6335  , p=0.0043  , df_denom=82, df_num=1\n",
      "ssr based chi2 test:   chi2=8.9494  , p=0.0028  , df=1\n",
      "likelihood ratio test: chi2=8.5089  , p=0.0035  , df=1\n",
      "parameter F test:         F=8.6335  , p=0.0043  , df_denom=82, df_num=1\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 2\n",
      "ssr based F test:         F=7.0861  , p=0.0015  , df_denom=79, df_num=2\n",
      "ssr based chi2 test:   chi2=15.0691 , p=0.0005  , df=2\n",
      "likelihood ratio test: chi2=13.8601 , p=0.0010  , df=2\n",
      "parameter F test:         F=7.0861  , p=0.0015  , df_denom=79, df_num=2\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 3\n",
      "ssr based F test:         F=5.3716  , p=0.0021  , df_denom=76, df_num=3\n",
      "ssr based chi2 test:   chi2=17.5989 , p=0.0005  , df=3\n",
      "likelihood ratio test: chi2=15.9610 , p=0.0012  , df=3\n",
      "parameter F test:         F=5.3716  , p=0.0021  , df_denom=76, df_num=3\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 4\n",
      "ssr based F test:         F=4.2670  , p=0.0037  , df_denom=73, df_num=4\n",
      "ssr based chi2 test:   chi2=19.1723 , p=0.0007  , df=4\n",
      "likelihood ratio test: chi2=17.2286 , p=0.0017  , df=4\n",
      "parameter F test:         F=4.2670  , p=0.0037  , df_denom=73, df_num=4\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{1: ({'ssr_ftest': (8.633545885637071, 0.004284464554853374, 82.0, 1),\n",
       "   'ssr_chi2test': (8.949407320477452, 0.0027755962266073856, 1),\n",
       "   'lrtest': (8.5089386758535, 0.0035340612493616388, 1),\n",
       "   'params_ftest': (8.633545885637142, 0.00428446455485323, 82.0, 1.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700219460>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277003f2dc0>,\n",
       "   array([[0., 1., 0.]])]),\n",
       " 2: ({'ssr_ftest': (7.086066484083583, 0.0014773099150423071, 79.0, 2),\n",
       "   'ssr_chi2test': (15.069103409190403, 0.0005343007343452818, 2),\n",
       "   'lrtest': (13.860069050258858, 0.0009779671033717059, 2),\n",
       "   'params_ftest': (7.086066484083597, 0.001477309915042292, 79.0, 2.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d712550>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d9411f0>,\n",
       "   array([[0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 1., 0.]])]),\n",
       " 3: ({'ssr_ftest': (5.371563589751934, 0.0020830250860398815, 76.0, 3),\n",
       "   'ssr_chi2test': (17.5989386032662, 0.0005320775305513886, 3),\n",
       "   'lrtest': (15.960991229451992, 0.0011550585779572268, 3),\n",
       "   'params_ftest': (5.371563589751955, 0.0020830250860398246, 76.0, 3.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700219b80>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700219730>,\n",
       "   array([[0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 4: ({'ssr_ftest': (4.2669986766780506, 0.0036983733142117604, 73.0, 4),\n",
       "   'ssr_chi2test': (19.172268026717816, 0.0007269964376486054, 4),\n",
       "   'lrtest': (17.228646135687484, 0.0017448806954374458, 4),\n",
       "   'params_ftest': (4.266998676677935, 0.003698373314212375, 73.0, 4.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700219eb0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700219400>,\n",
       "   array([[0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0.]])])}"
      ]
     },
     "execution_count": 770,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grangercausalitytests(df_during[['log_caution','log_new_case']],maxlag=4)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 803,
   "id": "1a526844",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 1\n",
      "ssr based F test:         F=3.1074  , p=0.0817  , df_denom=82, df_num=1\n",
      "ssr based chi2 test:   chi2=3.2211  , p=0.0727  , df=1\n",
      "likelihood ratio test: chi2=3.1616  , p=0.0754  , df=1\n",
      "parameter F test:         F=3.1074  , p=0.0817  , df_denom=82, df_num=1\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 2\n",
      "ssr based F test:         F=4.8639  , p=0.0102  , df_denom=79, df_num=2\n",
      "ssr based chi2 test:   chi2=10.3435 , p=0.0057  , df=2\n",
      "likelihood ratio test: chi2=9.7545  , p=0.0076  , df=2\n",
      "parameter F test:         F=4.8639  , p=0.0102  , df_denom=79, df_num=2\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 3\n",
      "ssr based F test:         F=2.6596  , p=0.0542  , df_denom=76, df_num=3\n",
      "ssr based chi2 test:   chi2=8.7137  , p=0.0334  , df=3\n",
      "likelihood ratio test: chi2=8.2860  , p=0.0405  , df=3\n",
      "parameter F test:         F=2.6596  , p=0.0542  , df_denom=76, df_num=3\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 4\n",
      "ssr based F test:         F=2.1016  , p=0.0893  , df_denom=73, df_num=4\n",
      "ssr based chi2 test:   chi2=9.4428  , p=0.0509  , df=4\n",
      "likelihood ratio test: chi2=8.9375  , p=0.0627  , df=4\n",
      "parameter F test:         F=2.1016  , p=0.0893  , df_denom=73, df_num=4\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{1: ({'ssr_ftest': (3.1074173325548085, 0.08166235402894213, 82.0, 1),\n",
       "   'ssr_chi2test': (3.2211033325263254, 0.07269461631338181, 1),\n",
       "   'lrtest': (3.161570300467247, 0.07539063294282126, 1),\n",
       "   'params_ftest': (3.1074173325547596, 0.08166235402894452, 82.0, 1.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27702615670>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c6811f0>,\n",
       "   array([[0., 1., 0.]])]),\n",
       " 2: ({'ssr_ftest': (4.863891163322425, 0.010183916871812616, 79.0, 2),\n",
       "   'ssr_chi2test': (10.343464752381864, 0.005674729532675084, 2),\n",
       "   'lrtest': (9.754517183607334, 0.007617869117196181, 2),\n",
       "   'params_ftest': (4.863891163322467, 0.010183916871812216, 79.0, 2.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c2d1c70>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c681e50>,\n",
       "   array([[0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 1., 0.]])]),\n",
       " 3: ({'ssr_ftest': (2.659599683205894, 0.054155721150439734, 76.0, 3),\n",
       "   'ssr_chi2test': (8.713688435766677, 0.0333500028717915, 3),\n",
       "   'lrtest': (8.285975858041695, 0.04045679030884618, 3),\n",
       "   'params_ftest': (2.659599683205896, 0.05415572115043946, 76.0, 3.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c681d90>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277009226d0>,\n",
       "   array([[0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 4: ({'ssr_ftest': (2.1015950705151836, 0.08926725539056625, 73.0, 4),\n",
       "   'ssr_chi2test': (9.442783330533974, 0.050936292624266126, 4),\n",
       "   'lrtest': (8.937525271816241, 0.06268016342410793, 4),\n",
       "   'params_ftest': (2.1015950705152133, 0.08926725539056281, 73.0, 4.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277009223a0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277009221c0>,\n",
       "   array([[0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0.]])])}"
      ]
     },
     "execution_count": 803,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grangercausalitytests(df_during[['log_new_case','log_caution']],maxlag=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 804,
   "id": "b9f00aa1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-2.0142191554131683,\n",
       " 0.5211792062490717,\n",
       " array([-4.0299263 , -3.40895789, -3.09472294]))"
      ]
     },
     "execution_count": 804,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = coint(df_during.new_case, df_during.caution)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 805,
   "id": "10c53e79",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-4.125218623036076,\n",
       " 0.004721947265089698,\n",
       " array([-4.0299263 , -3.40895789, -3.09472294]))"
      ]
     },
     "execution_count": 805,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = coint(df_during.new_case, df_during.caution2)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 807,
   "id": "cb9d25ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'New case')"
      ]
     },
     "execution_count": 807,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_lag, (ax1, ax2) = plt.subplots(1,2,figsize= (10,4))\n",
    "\n",
    "pd.plotting.lag_plot(df_during.caution2,ax=ax1)\n",
    "ax1.set_title('Detrended caution')\n",
    "pd.plotting.lag_plot(df_during.new_case,ax=ax2)\n",
    "ax2.set_title('New case')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 808,
   "id": "11317a10",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Log(new cases)')"
      ]
     },
     "execution_count": 808,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_lag_log, (ax1, ax2) = plt.subplots(1,2,figsize= (10,4))\n",
    "\n",
    "pd.plotting.lag_plot(df_during.log_caution,ax=ax1)\n",
    "ax1.set_title('Log(caution)')\n",
    "pd.plotting.lag_plot(df_during.log_caution,ax=ax2)\n",
    "ax2.set_title('Log(new cases)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 809,
   "id": "350f512d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_acf, (ax1, ax2) = plt.subplots(1,2, figsize=(12.5,4))\n",
    "\n",
    "plot_acf(df_during.log_caution, lags = 30, ax = ax1)\n",
    "ax1.set_title('Log(caution) ACF')\n",
    "plot_acf(df_during.log_new_case, lags = 30, ax = ax2)\n",
    "ax2.set_title('Log(new cases) ACF')\n",
    "plt.show()\n",
    "\n",
    "fig_pacf, (ax1, ax2) = plt.subplots(1,2, figsize=(12.5,4))\n",
    "\n",
    "plot_pacf(df_during.log_caution, lags = 30, ax = ax1, method='ywm')\n",
    "ax1.set_title('Log(caution) PACF')\n",
    "plot_pacf(df_during.log_new_case, lags = 30, ax = ax2, method='ywm')\n",
    "ax2.set_title('Log(new cases) PACF')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 810,
   "id": "048c7f49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Johansen cointegration test using maximum eigenvalue test statistic with 5% significance level</caption>\n",
       "<tr>\n",
       "  <th>r_0</th> <th>r_1</th> <th>test statistic</th> <th>critical value</th>\n",
       "</tr>\n",
       "<tr>\n",
       "    <td>0</td>   <td>1</td>          <td>13.61</td>          <td>14.26</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.table.SimpleTable'>"
      ]
     },
     "execution_count": 810,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sub = df_during[['new_case','caution2']]\n",
    "train, test = df_sub[0:round(0.8*(len(df_sub.caution2)))], df_sub[round(0.8*(len(df_sub.caution2))):]\n",
    "\n",
    "rank = select_coint_rank(train, det_order=0, k_ar_diff=4, method='maxeig', signif=0.05)\n",
    "\n",
    "rank.summary()\n",
    "# not quite\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 813,
   "id": "9c52fe82",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:536: ValueWarning: No frequency information was provided, so inferred frequency W-SUN will be used.\n",
      "  warnings.warn('No frequency information was'\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Det. terms outside the coint. relation & lagged endog. parameters for equation new_case</caption>\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>L1.new_case</th> <td>    0.5265</td> <td>    0.118</td> <td>    4.454</td> <td> 0.000</td> <td>    0.295</td> <td>    0.758</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.caution2</th> <td>    0.3892</td> <td>    0.204</td> <td>    1.905</td> <td> 0.057</td> <td>   -0.011</td> <td>    0.790</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.new_case</th> <td>    0.0630</td> <td>    0.120</td> <td>    0.526</td> <td> 0.599</td> <td>   -0.172</td> <td>    0.298</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.caution2</th> <td>    0.3695</td> <td>    0.272</td> <td>    1.359</td> <td> 0.174</td> <td>   -0.164</td> <td>    0.903</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L3.new_case</th> <td>   -0.0507</td> <td>    0.120</td> <td>   -0.423</td> <td> 0.672</td> <td>   -0.286</td> <td>    0.184</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L3.caution2</th> <td>   -0.9005</td> <td>    0.269</td> <td>   -3.353</td> <td> 0.001</td> <td>   -1.427</td> <td>   -0.374</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L4.new_case</th> <td>    0.1104</td> <td>    0.108</td> <td>    1.021</td> <td> 0.307</td> <td>   -0.102</td> <td>    0.322</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L4.caution2</th> <td>    0.3862</td> <td>    0.214</td> <td>    1.803</td> <td> 0.071</td> <td>   -0.034</td> <td>    0.806</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Det. terms outside the coint. relation & lagged endog. parameters for equation caution2</caption>\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>L1.new_case</th> <td>    0.0301</td> <td>    0.073</td> <td>    0.413</td> <td> 0.679</td> <td>   -0.113</td> <td>    0.173</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.caution2</th> <td>    1.0678</td> <td>    0.126</td> <td>    8.480</td> <td> 0.000</td> <td>    0.821</td> <td>    1.315</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.new_case</th> <td>    0.1052</td> <td>    0.074</td> <td>    1.427</td> <td> 0.154</td> <td>   -0.039</td> <td>    0.250</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.caution2</th> <td>   -0.0308</td> <td>    0.168</td> <td>   -0.184</td> <td> 0.854</td> <td>   -0.359</td> <td>    0.298</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L3.new_case</th> <td>    0.0864</td> <td>    0.074</td> <td>    1.170</td> <td> 0.242</td> <td>   -0.058</td> <td>    0.231</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L3.caution2</th> <td>   -0.5648</td> <td>    0.165</td> <td>   -3.413</td> <td> 0.001</td> <td>   -0.889</td> <td>   -0.240</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L4.new_case</th> <td>   -0.0199</td> <td>    0.067</td> <td>   -0.299</td> <td> 0.765</td> <td>   -0.151</td> <td>    0.111</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L4.caution2</th> <td>    0.1394</td> <td>    0.132</td> <td>    1.056</td> <td> 0.291</td> <td>   -0.119</td> <td>    0.398</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Loading coefficients (alpha) for equation new_case</caption>\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>ec1</th> <td>   -0.1514</td> <td>    0.055</td> <td>   -2.748</td> <td> 0.006</td> <td>   -0.259</td> <td>   -0.043</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Loading coefficients (alpha) for equation caution2</caption>\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>ec1</th> <td>    0.0110</td> <td>    0.034</td> <td>    0.324</td> <td> 0.746</td> <td>   -0.056</td> <td>    0.078</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Cointegration relations for loading-coefficients-column 1</caption>\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>beta.1</th> <td>    1.0000</td> <td>        0</td> <td>        0</td> <td> 0.000</td> <td>    1.000</td> <td>    1.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>beta.2</th> <td>   -0.7032</td> <td>    0.081</td> <td>   -8.695</td> <td> 0.000</td> <td>   -0.862</td> <td>   -0.545</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "Det. terms outside the coint. relation & lagged endog. parameters for equation new_case\n",
       "===============================================================================\n",
       "                  coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------\n",
       "L1.new_case     0.5265      0.118      4.454      0.000       0.295       0.758\n",
       "L1.caution2     0.3892      0.204      1.905      0.057      -0.011       0.790\n",
       "L2.new_case     0.0630      0.120      0.526      0.599      -0.172       0.298\n",
       "L2.caution2     0.3695      0.272      1.359      0.174      -0.164       0.903\n",
       "L3.new_case    -0.0507      0.120     -0.423      0.672      -0.286       0.184\n",
       "L3.caution2    -0.9005      0.269     -3.353      0.001      -1.427      -0.374\n",
       "L4.new_case     0.1104      0.108      1.021      0.307      -0.102       0.322\n",
       "L4.caution2     0.3862      0.214      1.803      0.071      -0.034       0.806\n",
       "Det. terms outside the coint. relation & lagged endog. parameters for equation caution2\n",
       "===============================================================================\n",
       "                  coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-------------------------------------------------------------------------------\n",
       "L1.new_case     0.0301      0.073      0.413      0.679      -0.113       0.173\n",
       "L1.caution2     1.0678      0.126      8.480      0.000       0.821       1.315\n",
       "L2.new_case     0.1052      0.074      1.427      0.154      -0.039       0.250\n",
       "L2.caution2    -0.0308      0.168     -0.184      0.854      -0.359       0.298\n",
       "L3.new_case     0.0864      0.074      1.170      0.242      -0.058       0.231\n",
       "L3.caution2    -0.5648      0.165     -3.413      0.001      -0.889      -0.240\n",
       "L4.new_case    -0.0199      0.067     -0.299      0.765      -0.151       0.111\n",
       "L4.caution2     0.1394      0.132      1.056      0.291      -0.119       0.398\n",
       "              Loading coefficients (alpha) for equation new_case              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ec1           -0.1514      0.055     -2.748      0.006      -0.259      -0.043\n",
       "              Loading coefficients (alpha) for equation caution2              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ec1            0.0110      0.034      0.324      0.746      -0.056       0.078\n",
       "          Cointegration relations for loading-coefficients-column 1           \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "beta.1         1.0000          0          0      0.000       1.000       1.000\n",
       "beta.2        -0.7032      0.081     -8.695      0.000      -0.862      -0.545\n",
       "==============================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 813,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_VECM = VECM(train, coint_rank=1, k_ar_diff=4)\n",
    "vecm_res = model_VECM.fit()\n",
    "\n",
    "fc_vecm, lower, upper = vecm_res.predict(steps=len(test), alpha=0.05)\n",
    "\n",
    "plt.plot(train.new_case, color='dimgray')\n",
    "plt.plot(test.new_case, color='royalblue')\n",
    "plt.plot(test.index,fc_vecm[:,0], color='orangered',linestyle='--')\n",
    "plt.xticks(rotation=45)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('New cases (normalized)')\n",
    "plt.legend(['Train','Test actual','Test predicted'])\n",
    "plt.show()\n",
    "\n",
    "vecm_res.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 814,
   "id": "c3dd961f",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "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": [
    "fig,ax = plt.subplots()\n",
    "\n",
    "ax.plot(train.caution2,color='dimgray')\n",
    "ax.plot(test.caution2,color='royalblue')\n",
    "ax.plot(test.index,fc_vecm[:,1],color='orangered',linestyle='--')\n",
    "\n",
    "ax.set_ylabel('Normalized Caution (detrended)')\n",
    "ax.set_xlabel('Date')\n",
    "ax.legend(['Train','Test actual','Test predicted'])\n",
    "\n",
    "plt.xticks(rotation=45)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 815,
   "id": "17dcfe7c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create date index for next 6 weeks\n",
    "\n",
    "L=6\n",
    "i=0\n",
    "fc_ind = [None]*(L+1)\n",
    "fc_ind[0]=df_sub.index[-1]\n",
    "while i<L:\n",
    "    fc_ind[i+1] = fc_ind[i] + pd.Timedelta(days=7)\n",
    "    i+=1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 816,
   "id": "bec7472d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:536: ValueWarning: No frequency information was provided, so inferred frequency W-SUN will be used.\n",
      "  warnings.warn('No frequency information was'\n"
     ]
    },
    {
     "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": [
    "model_VECM = VECM(df_sub, coint_rank=1, k_ar_diff=4, dates=df_sub.index)\n",
    "vecm_res = model_VECM.fit()\n",
    "\n",
    "fc_vecm, lower, upper = vecm_res.predict(steps=L, alpha=0.05)\n",
    "\n",
    "plt.plot(df_sub.new_case,color='dimgray')\n",
    "plt.plot(fc_ind,np.insert(fc_vecm[:,0],0,df_sub.new_case.values[-1]),color='orangered',linestyle=':')\n",
    "\n",
    "plt.fill_between(fc_ind,\n",
    "                 np.insert(lower[:,0],0,df_sub.new_case.values[-1]),\n",
    "                 np.insert(upper[:,0],0,df_sub.new_case.values[-1]),\n",
    "                 color='orangered',alpha=0.2)\n",
    "\n",
    "plt.ylabel('New cases')\n",
    "plt.xlabel('Date')\n",
    "plt.title('Prediction for next 6 weeks')\n",
    "\n",
    "plt.xticks(rotation=45)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 817,
   "id": "043e5b42",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean VECM new case 6 week MAPE error: 10.065328574128229\n",
      "Mean Naive new case 6 week MAPE error: 16.000932197597596\n",
      "Mean VECM caution 6 week MAPE error: 7.661977147557477\n",
      "Mean Naive caution 6 week MAPE error: 10.070064243323605\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:536: ValueWarning: No frequency information was provided, so inferred frequency W-SUN will be used.\n",
      "  warnings.warn('No frequency information was'\n",
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:536: ValueWarning: No frequency information was provided, so inferred frequency W-SUN will be used.\n",
      "  warnings.warn('No frequency information was'\n",
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:536: ValueWarning: No frequency information was provided, so inferred frequency W-SUN will be used.\n",
      "  warnings.warn('No frequency information was'\n"
     ]
    }
   ],
   "source": [
    "err_nc = [None]*(3)\n",
    "err_nc_naive = [None]*3\n",
    "err_cau = [None]*(3)\n",
    "err_cau_naive = [None]*3\n",
    "fc_weeks = 6\n",
    "new_case_naive = [None]*fc_weeks\n",
    "i = 0\n",
    "\n",
    "for frac_train in np.arange(0.8,0.95, 0.05):\n",
    "    train, test = df_sub[0:round(frac_train*(len(df_sub.caution2)))], df_sub[round(frac_train*(len(df_sub.caution2))):]\n",
    "    \n",
    "    model_VECM = VECM(train, coint_rank=1, k_ar_diff=4)\n",
    "    vecm_res = model_VECM.fit()\n",
    "\n",
    "    fc_vecm = vecm_res.predict(steps=fc_weeks)\n",
    "    \n",
    "    err_nc[i] = mape(test.new_case.values[:fc_weeks], fc_vecm[:,0])\n",
    "    new_case_naive[0:fc_weeks] = np.repeat(test.new_case.values[0],fc_weeks)\n",
    "    err_nc_naive[i] = mape(test.new_case.values[:fc_weeks],new_case_naive)\n",
    "    \n",
    "    err_cau[i] = mape(test.caution2.values[:fc_weeks], fc_vecm[:,1])\n",
    "    new_case_naive[0:fc_weeks] = np.repeat(test.caution2.values[0],fc_weeks)\n",
    "    err_cau_naive[i] = mape(test.caution2.values[:fc_weeks],new_case_naive)\n",
    "    \n",
    "    i+=1\n",
    "    \n",
    "print('Mean VECM new case',str(fc_weeks),'week MAPE error:', np.mean(err_nc))\n",
    "print('Mean Naive new case',str(fc_weeks),'week MAPE error:', np.mean(err_nc_naive))\n",
    "\n",
    "print('Mean VECM caution',str(fc_weeks),'week MAPE error:', np.mean(err_cau))\n",
    "print('Mean Naive caution',str(fc_weeks),'week MAPE error:', np.mean(err_cau_naive))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 818,
   "id": "6ac1190e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pmdarima\\arima\\_auto_solvers.py:103: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
      "  gen = random_state.permutation(list(gen))[:n_fits]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ARIMA(3,0,0)(0,0,0)[0]             : AIC=37.164, Time=0.18 sec\n",
      " ARIMA(1,0,4)(0,0,0)[0]             : AIC=42.158, Time=0.07 sec\n",
      " ARIMA(0,0,3)(0,0,0)[0]             : AIC=40.038, Time=0.05 sec\n",
      " ARIMA(4,0,0)(0,0,0)[0]             : AIC=38.295, Time=0.06 sec\n",
      " ARIMA(3,0,2)(0,0,0)[0]             : AIC=158.461, Time=0.08 sec\n",
      " ARIMA(2,0,3)(0,0,0)[0]             : AIC=38.602, Time=0.06 sec\n",
      " ARIMA(3,0,1)(0,0,0)[0]             : AIC=38.909, Time=0.06 sec\n",
      " ARIMA(0,0,4)(0,0,0)[0]             : AIC=42.419, Time=0.06 sec\n",
      " ARIMA(0,0,2)(0,0,0)[0]             : AIC=47.447, Time=0.07 sec\n",
      " ARIMA(1,0,0)(0,0,0)[0]             : AIC=67.356, Time=0.05 sec\n",
      " ARIMA(1,0,1)(0,0,0)[0]             : AIC=41.872, Time=0.06 sec\n",
      " ARIMA(0,0,1)(0,0,0)[0]             : AIC=129.242, Time=0.05 sec\n",
      " ARIMA(2,0,2)(0,0,0)[0]             : AIC=37.944, Time=0.07 sec\n",
      " ARIMA(0,0,0)(0,0,0)[0]             : AIC=148.636, Time=0.06 sec\n",
      " ARIMA(5,0,0)(0,0,0)[0]             : AIC=39.477, Time=0.07 sec\n",
      " ARIMA(4,0,1)(0,0,0)[0]             : AIC=41.386, Time=0.06 sec\n",
      " ARIMA(1,0,3)(0,0,0)[0]             : AIC=41.692, Time=0.09 sec\n",
      " ARIMA(1,0,2)(0,0,0)[0]             : AIC=39.405, Time=0.07 sec\n",
      " ARIMA(2,0,1)(0,0,0)[0]             : AIC=37.022, Time=0.05 sec\n",
      " ARIMA(2,0,0)(0,0,0)[0]             : AIC=35.609, Time=0.09 sec\n",
      " ARIMA(0,0,5)(0,0,0)[0]             : AIC=44.867, Time=0.11 sec\n",
      "\n",
      "Best model:  ARIMA(2,0,0)(0,0,0)[0]          \n",
      "Total fit time: 1.546 seconds\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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>69</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>SARIMAX(2, 0, 0)</td> <th>  Log Likelihood     </th> <td>-18.446</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Sat, 11 Dec 2021</td> <th>  AIC                </th> <td>44.893</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>19:23:09</td>     <th>  BIC                </th> <td>53.829</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                  <td>0</td>        <th>  HQIC               </th> <td>48.438</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                       <td> - 69</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>x1</th>     <td>    0.6172</td> <td>    0.048</td> <td>   12.813</td> <td> 0.000</td> <td>    0.523</td> <td>    0.712</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1</th>  <td>    1.2373</td> <td>    0.078</td> <td>   15.959</td> <td> 0.000</td> <td>    1.085</td> <td>    1.389</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L2</th>  <td>   -0.5636</td> <td>    0.076</td> <td>   -7.455</td> <td> 0.000</td> <td>   -0.712</td> <td>   -0.415</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    0.0974</td> <td>    0.010</td> <td>    9.865</td> <td> 0.000</td> <td>    0.078</td> <td>    0.117</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.16</td> <th>  Jarque-Bera (JB):  </th> <td>147.25</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.69</td> <th>  Prob(JB):          </th>  <td>0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>1.45</td> <th>  Skew:              </th>  <td>0.02</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.38</td> <th>  Kurtosis:          </th>  <td>10.16</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:                   69\n",
       "Model:               SARIMAX(2, 0, 0)   Log Likelihood                 -18.446\n",
       "Date:                Sat, 11 Dec 2021   AIC                             44.893\n",
       "Time:                        19:23:09   BIC                             53.829\n",
       "Sample:                             0   HQIC                            48.438\n",
       "                                 - 69                                         \n",
       "Covariance Type:                  opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "x1             0.6172      0.048     12.813      0.000       0.523       0.712\n",
       "ar.L1          1.2373      0.078     15.959      0.000       1.085       1.389\n",
       "ar.L2         -0.5636      0.076     -7.455      0.000      -0.712      -0.415\n",
       "sigma2         0.0974      0.010      9.865      0.000       0.078       0.117\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   0.16   Jarque-Bera (JB):               147.25\n",
       "Prob(Q):                              0.69   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               1.45   Skew:                             0.02\n",
       "Prob(H) (two-sided):                  0.38   Kurtosis:                        10.16\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 818,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_log = df_during[['log_new_case','log_caution']]\n",
    "train, test = df_log[0:round(0.8*(len(df_during.log_caution)))], df_log[round(0.8*(len(df_during.log_caution))):]\n",
    "\n",
    "model = auto_arima(df_during.log_caution,start_p=2, d=0, q=0, seasonal=False, trace=1, \n",
    "                   exogenous=np.reshape(df_during.log_new_case.values,(-1,1)),\n",
    "                   stepwise=False, random=True, n_fits=50, random_state=1, method='nm')\n",
    "\n",
    "model.fit(train.log_caution, exogenous=np.reshape(train.log_new_case.values,(-1,1)))\n",
    "forecast = model.predict(exogenous=np.reshape(test.log_new_case.values,(-1,1)), n_periods=len(test))\n",
    "\n",
    "forecast = pd.DataFrame(forecast,index = test.index,columns=['log_caution_pred'])\n",
    "forecast = pd.concat([test,forecast],axis=1)\n",
    "plt.plot(np.exp(train.log_caution)-epsilon)\n",
    "plt.plot(np.exp(forecast.log_caution)-epsilon)\n",
    "plt.plot(np.exp(forecast.log_caution_pred)-epsilon, color='red',linestyle='--')\n",
    "plt.xticks(rotation=45)\n",
    "plt.ylabel('Normalized caution (detrended)')\n",
    "plt.legend(['Train','Test actual','Test predicted'])\n",
    "plt.show()\n",
    "\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 819,
   "id": "1862a0ff",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pmdarima\\arima\\_auto_solvers.py:103: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
      "  gen = random_state.permutation(list(gen))[:n_fits]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ARIMA(3,0,0)(0,0,0)[0]             : AIC=37.164, Time=0.10 sec\n",
      " ARIMA(1,0,4)(0,0,0)[0]             : AIC=42.158, Time=0.07 sec\n",
      " ARIMA(0,0,3)(0,0,0)[0]             : AIC=40.038, Time=0.06 sec\n",
      " ARIMA(4,0,0)(0,0,0)[0]             : AIC=38.295, Time=0.07 sec\n",
      " ARIMA(3,0,2)(0,0,0)[0]             : AIC=158.461, Time=0.07 sec\n",
      " ARIMA(2,0,3)(0,0,0)[0]             : AIC=38.602, Time=0.05 sec\n",
      " ARIMA(3,0,1)(0,0,0)[0]             : AIC=38.909, Time=0.06 sec\n",
      " ARIMA(0,0,4)(0,0,0)[0]             : AIC=42.419, Time=0.07 sec\n",
      " ARIMA(0,0,2)(0,0,0)[0]             : AIC=47.447, Time=0.05 sec\n",
      " ARIMA(1,0,0)(0,0,0)[0]             : AIC=67.356, Time=0.04 sec\n",
      " ARIMA(1,0,1)(0,0,0)[0]             : AIC=41.872, Time=0.05 sec\n",
      " ARIMA(0,0,1)(0,0,0)[0]             : AIC=129.242, Time=0.05 sec\n",
      " ARIMA(2,0,2)(0,0,0)[0]             : AIC=37.944, Time=0.07 sec\n",
      " ARIMA(0,0,0)(0,0,0)[0]             : AIC=148.636, Time=0.04 sec\n",
      " ARIMA(5,0,0)(0,0,0)[0]             : AIC=39.477, Time=0.06 sec\n",
      " ARIMA(4,0,1)(0,0,0)[0]             : AIC=41.386, Time=0.06 sec\n",
      " ARIMA(1,0,3)(0,0,0)[0]             : AIC=41.692, Time=0.06 sec\n",
      " ARIMA(1,0,2)(0,0,0)[0]             : AIC=39.405, Time=0.07 sec\n",
      " ARIMA(2,0,1)(0,0,0)[0]             : AIC=37.022, Time=0.05 sec\n",
      " ARIMA(2,0,0)(0,0,0)[0]             : AIC=35.609, Time=0.05 sec\n",
      " ARIMA(0,0,5)(0,0,0)[0]             : AIC=44.867, Time=0.09 sec\n",
      "\n",
      "Best model:  ARIMA(2,0,0)(0,0,0)[0]          \n",
      "Total fit time: 1.327 seconds\n",
      " ARIMA(3,0,0)(0,0,0)[0]             : AIC=37.164, Time=0.10 sec\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pmdarima\\arima\\_auto_solvers.py:103: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
      "  gen = random_state.permutation(list(gen))[:n_fits]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ARIMA(1,0,4)(0,0,0)[0]             : AIC=42.158, Time=0.13 sec\n",
      " ARIMA(0,0,3)(0,0,0)[0]             : AIC=40.038, Time=0.08 sec\n",
      " ARIMA(4,0,0)(0,0,0)[0]             : AIC=38.295, Time=0.08 sec\n",
      " ARIMA(3,0,2)(0,0,0)[0]             : AIC=158.461, Time=0.07 sec\n",
      " ARIMA(2,0,3)(0,0,0)[0]             : AIC=38.602, Time=0.08 sec\n",
      " ARIMA(3,0,1)(0,0,0)[0]             : AIC=38.909, Time=0.06 sec\n",
      " ARIMA(0,0,4)(0,0,0)[0]             : AIC=42.419, Time=0.06 sec\n",
      " ARIMA(0,0,2)(0,0,0)[0]             : AIC=47.447, Time=0.07 sec\n",
      " ARIMA(1,0,0)(0,0,0)[0]             : AIC=67.356, Time=0.04 sec\n",
      " ARIMA(1,0,1)(0,0,0)[0]             : AIC=41.872, Time=0.05 sec\n",
      " ARIMA(0,0,1)(0,0,0)[0]             : AIC=129.242, Time=0.05 sec\n",
      " ARIMA(2,0,2)(0,0,0)[0]             : AIC=37.944, Time=0.06 sec\n",
      " ARIMA(0,0,0)(0,0,0)[0]             : AIC=148.636, Time=0.04 sec\n",
      " ARIMA(5,0,0)(0,0,0)[0]             : AIC=39.477, Time=0.07 sec\n",
      " ARIMA(4,0,1)(0,0,0)[0]             : AIC=41.386, Time=0.07 sec\n",
      " ARIMA(1,0,3)(0,0,0)[0]             : AIC=41.692, Time=0.07 sec\n",
      " ARIMA(1,0,2)(0,0,0)[0]             : AIC=39.405, Time=0.06 sec\n",
      " ARIMA(2,0,1)(0,0,0)[0]             : AIC=37.022, Time=0.05 sec\n",
      " ARIMA(2,0,0)(0,0,0)[0]             : AIC=35.609, Time=0.10 sec\n",
      " ARIMA(0,0,5)(0,0,0)[0]             : AIC=44.867, Time=0.10 sec\n",
      "\n",
      "Best model:  ARIMA(2,0,0)(0,0,0)[0]          \n",
      "Total fit time: 1.511 seconds\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pmdarima\\arima\\_auto_solvers.py:103: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
      "  gen = random_state.permutation(list(gen))[:n_fits]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ARIMA(3,0,0)(0,0,0)[0]             : AIC=37.164, Time=0.14 sec\n",
      " ARIMA(1,0,4)(0,0,0)[0]             : AIC=42.158, Time=0.10 sec\n",
      " ARIMA(0,0,3)(0,0,0)[0]             : AIC=40.038, Time=0.12 sec\n",
      " ARIMA(4,0,0)(0,0,0)[0]             : AIC=38.295, Time=0.11 sec\n",
      " ARIMA(3,0,2)(0,0,0)[0]             : AIC=158.461, Time=0.09 sec\n",
      " ARIMA(2,0,3)(0,0,0)[0]             : AIC=38.602, Time=0.09 sec\n",
      " ARIMA(3,0,1)(0,0,0)[0]             : AIC=38.909, Time=0.11 sec\n",
      " ARIMA(0,0,4)(0,0,0)[0]             : AIC=42.419, Time=0.10 sec\n",
      " ARIMA(0,0,2)(0,0,0)[0]             : AIC=47.447, Time=0.05 sec\n",
      " ARIMA(1,0,0)(0,0,0)[0]             : AIC=67.356, Time=0.04 sec\n",
      " ARIMA(1,0,1)(0,0,0)[0]             : AIC=41.872, Time=0.10 sec\n",
      " ARIMA(0,0,1)(0,0,0)[0]             : AIC=129.242, Time=0.09 sec\n",
      " ARIMA(2,0,2)(0,0,0)[0]             : AIC=37.944, Time=0.07 sec\n",
      " ARIMA(0,0,0)(0,0,0)[0]             : AIC=148.636, Time=0.06 sec\n",
      " ARIMA(5,0,0)(0,0,0)[0]             : AIC=39.477, Time=0.11 sec\n",
      " ARIMA(4,0,1)(0,0,0)[0]             : AIC=41.386, Time=0.10 sec\n",
      " ARIMA(1,0,3)(0,0,0)[0]             : AIC=41.692, Time=0.14 sec\n",
      " ARIMA(1,0,2)(0,0,0)[0]             : AIC=39.405, Time=0.07 sec\n",
      " ARIMA(2,0,1)(0,0,0)[0]             : AIC=37.022, Time=0.08 sec\n",
      " ARIMA(2,0,0)(0,0,0)[0]             : AIC=35.609, Time=0.05 sec\n",
      " ARIMA(0,0,5)(0,0,0)[0]             : AIC=44.867, Time=0.05 sec\n",
      "\n",
      "Best model:  ARIMA(2,0,0)(0,0,0)[0]          \n",
      "Total fit time: 1.908 seconds\n",
      "Mean ARMAX 6 week error: 7.6426643787496475\n",
      "Mean Naive 6 week error: 10.070064243323605\n"
     ]
    }
   ],
   "source": [
    "err = [None]*(3)\n",
    "err_naive = [None]*3\n",
    "fc_weeks = 6\n",
    "new_case_naive = [None]*fc_weeks\n",
    "i = 0\n",
    "\n",
    "for frac_train in np.arange(0.8,0.95, 0.05):    \n",
    "    df_log = df_during[['log_new_case','log_caution']]\n",
    "    train, test = df_log[0:round(frac_train*(len(df_during.log_caution)))], df_log[round(frac_train*(len(df_during.log_caution))):]\n",
    "\n",
    "    model = auto_arima(df_during.log_caution,start_p=2, d=0, q=0, seasonal=False, trace=1, \n",
    "                   exogenous=np.reshape(df_during.log_new_case.values,(-1,1)),\n",
    "                   stepwise=False, random=True, n_fits=50, random_state=1, method='nm')\n",
    "    res = model.fit(train.log_caution, exogenous=np.reshape(train.log_new_case.values,(-1,1)))\n",
    "\n",
    "    fc = model.predict(exogenous=np.reshape(test.log_new_case.values[:fc_weeks],(-1,1)), n_periods=fc_weeks)\n",
    "    \n",
    "    err[i] = mape(np.exp(test.log_caution.values[:fc_weeks])-epsilon, np.exp(fc)-epsilon)\n",
    "    new_case_naive[0:fc_weeks] = np.repeat(test.log_caution.values[0],fc_weeks)\n",
    "    err_naive[i] = mape(np.exp(test.log_caution.values[:fc_weeks])-epsilon, np.exp(new_case_naive)-epsilon)\n",
    "    i+=1\n",
    "    \n",
    "print('Mean ARMAX',str(fc_weeks),'week error:', np.mean(err))\n",
    "print('Mean Naive',str(fc_weeks),'week error:', np.mean(err_naive))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 820,
   "id": "72ac55c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\statsmodels\\tsa\\base\\tsa_model.py:536: ValueWarning: No frequency information was provided, so inferred frequency W-SUN will be used.\n",
      "  warnings.warn('No frequency information was'\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Statespace Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>   <td>['log_new_case', 'log_caution']</td> <th>  No. Observations:  </th>   <td>69</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                       <td>VAR(2)</td>              <th>  Log Likelihood     </th> <td>10.530</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                           <td>+ intercept</td>           <th>  AIC                </th>  <td>4.939</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                   <td>Sat, 11 Dec 2021</td>         <th>  BIC                </th> <td>33.983</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                       <td>19:23:22</td>             <th>  HQIC               </th> <td>16.462</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                    <td>04-12-2020</td>            <th>                     </th>    <td> </td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                          <td>- 08-01-2021</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",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>2.15, 0.22</td> <th>  Jarque-Bera (JB):  </th> <td>0.51, 303.94</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.14, 0.64</td> <th>  Prob(JB):          </th>  <td>0.78, 0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>0.36, 2.36</td> <th>  Skew:              </th>  <td>-0.18, 1.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.02, 0.05</td> <th>  Kurtosis:          </th>  <td>3.23, 13.08</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Results for equation log_new_case</caption>\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>intercept</th>       <td>   -0.0700</td> <td>    0.060</td> <td>   -1.175</td> <td> 0.240</td> <td>   -0.187</td> <td>    0.047</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.log_new_case</th> <td>    1.6094</td> <td>    0.113</td> <td>   14.234</td> <td> 0.000</td> <td>    1.388</td> <td>    1.831</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.log_caution</th>  <td>    0.1738</td> <td>    0.114</td> <td>    1.526</td> <td> 0.127</td> <td>   -0.049</td> <td>    0.397</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.log_new_case</th> <td>   -0.6625</td> <td>    0.102</td> <td>   -6.500</td> <td> 0.000</td> <td>   -0.862</td> <td>   -0.463</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.log_caution</th>  <td>   -0.1493</td> <td>    0.074</td> <td>   -2.008</td> <td> 0.045</td> <td>   -0.295</td> <td>   -0.004</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Results for equation log_caution</caption>\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>intercept</th>       <td>   -0.0762</td> <td>    0.179</td> <td>   -0.425</td> <td> 0.671</td> <td>   -0.427</td> <td>    0.275</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.log_new_case</th> <td>    0.4288</td> <td>    0.179</td> <td>    2.400</td> <td> 0.016</td> <td>    0.079</td> <td>    0.779</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L1.log_caution</th>  <td>    1.3512</td> <td>    0.076</td> <td>   17.804</td> <td> 0.000</td> <td>    1.202</td> <td>    1.500</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.log_new_case</th> <td>   -0.2469</td> <td>    0.169</td> <td>   -1.462</td> <td> 0.144</td> <td>   -0.578</td> <td>    0.084</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>L2.log_caution</th>  <td>   -0.6866</td> <td>    0.078</td> <td>   -8.845</td> <td> 0.000</td> <td>   -0.839</td> <td>   -0.534</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Error covariance matrix</caption>\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>sqrt.var.log_new_case</th>             <td>    0.1751</td> <td>    0.015</td> <td>   11.847</td> <td> 0.000</td> <td>    0.146</td> <td>    0.204</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sqrt.cov.log_new_case.log_caution</th> <td>   -0.0309</td> <td>    0.075</td> <td>   -0.411</td> <td> 0.681</td> <td>   -0.178</td> <td>    0.117</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sqrt.var.log_caution</th>              <td>    0.2728</td> <td>    0.020</td> <td>   13.849</td> <td> 0.000</td> <td>    0.234</td> <td>    0.311</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",
       "                                  Statespace Model Results                                 \n",
       "===========================================================================================\n",
       "Dep. Variable:     ['log_new_case', 'log_caution']   No. Observations:                   69\n",
       "Model:                                      VAR(2)   Log Likelihood                  10.530\n",
       "                                       + intercept   AIC                              4.939\n",
       "Date:                             Sat, 11 Dec 2021   BIC                             33.983\n",
       "Time:                                     19:23:22   HQIC                            16.462\n",
       "Sample:                                 04-12-2020                                         \n",
       "                                      - 08-01-2021                                         \n",
       "Covariance Type:                               opg                                         \n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):             2.15, 0.22   Jarque-Bera (JB):         0.51, 303.94\n",
       "Prob(Q):                        0.14, 0.64   Prob(JB):                   0.78, 0.00\n",
       "Heteroskedasticity (H):         0.36, 2.36   Skew:                      -0.18, 1.02\n",
       "Prob(H) (two-sided):            0.02, 0.05   Kurtosis:                  3.23, 13.08\n",
       "                         Results for equation log_new_case                         \n",
       "===================================================================================\n",
       "                      coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-----------------------------------------------------------------------------------\n",
       "intercept          -0.0700      0.060     -1.175      0.240      -0.187       0.047\n",
       "L1.log_new_case     1.6094      0.113     14.234      0.000       1.388       1.831\n",
       "L1.log_caution      0.1738      0.114      1.526      0.127      -0.049       0.397\n",
       "L2.log_new_case    -0.6625      0.102     -6.500      0.000      -0.862      -0.463\n",
       "L2.log_caution     -0.1493      0.074     -2.008      0.045      -0.295      -0.004\n",
       "                          Results for equation log_caution                         \n",
       "===================================================================================\n",
       "                      coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-----------------------------------------------------------------------------------\n",
       "intercept          -0.0762      0.179     -0.425      0.671      -0.427       0.275\n",
       "L1.log_new_case     0.4288      0.179      2.400      0.016       0.079       0.779\n",
       "L1.log_caution      1.3512      0.076     17.804      0.000       1.202       1.500\n",
       "L2.log_new_case    -0.2469      0.169     -1.462      0.144      -0.578       0.084\n",
       "L2.log_caution     -0.6866      0.078     -8.845      0.000      -0.839      -0.534\n",
       "                                       Error covariance matrix                                       \n",
       "=====================================================================================================\n",
       "                                        coef    std err          z      P>|z|      [0.025      0.975]\n",
       "-----------------------------------------------------------------------------------------------------\n",
       "sqrt.var.log_new_case                 0.1751      0.015     11.847      0.000       0.146       0.204\n",
       "sqrt.cov.log_new_case.log_caution    -0.0309      0.075     -0.411      0.681      -0.178       0.117\n",
       "sqrt.var.log_caution                  0.2728      0.020     13.849      0.000       0.234       0.311\n",
       "=====================================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 820,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_during2 = df_during.dropna()     # creating this helps indices stay in sync\n",
    "df_log = df_during2[['log_new_case','log_caution']]\n",
    "train, test = df_log[0:round(0.8*(len(df_log.log_caution)))], df_during[round(0.8*(len(df_log.log_caution))):]\n",
    "temp_train, temp_test = df_during2.temp_C[0:round(0.8*(len(df_log.log_caution)))], df_during2.temp_C[round(0.8*(len(df_log.log_caution))):]\n",
    "\n",
    "#model2 = sm.tsa.VARMAX(endog=train, exog=temp_train, order=(2,0), trend='c')\n",
    "# Temperature coefficient wasn't significant for FL, seemed to overweight winter surge of 2020 for NY\n",
    "model2 = sm.tsa.VARMAX(endog=train, order = (2,0), trend='c')\n",
    "model2_fit = model2.fit(maxiter=5000, disp=True)\n",
    "\n",
    "# Only can forecast from end of sample\n",
    "fc = model2_fit.forecast(steps=len(test.log_new_case))\n",
    "\n",
    "fig,ax = plt.subplots()\n",
    "\n",
    "ax.plot(np.exp(train.log_new_case)-epsilon,color='black')\n",
    "\n",
    "ax.plot(np.exp(test.log_new_case)-epsilon,color='blue')\n",
    "ax.set_ylabel('New cases transformed',color='blue')\n",
    "ax.set_xlabel('Time')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "ax.plot(np.exp(fc.log_new_case)-epsilon,color='red',linestyle='--')\n",
    "plt.show()\n",
    "\n",
    "model2_fit.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 821,
   "id": "43f50d0e",
   "metadata": {},
   "outputs": [
    {
     "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": [
    "fig,ax = plt.subplots()\n",
    "\n",
    "ax.plot(np.exp(train.log_caution)-epsilon,color='black')\n",
    "\n",
    "ax.plot(np.exp(test.log_caution)-epsilon,color='blue')\n",
    "ax.set_ylabel('Relative caution',color='blue')\n",
    "ax.set_xlabel('Time')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "ax.plot(np.exp(fc.log_caution)-epsilon,color='red',linestyle='--')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 824,
   "id": "363680c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/1299464790.py:19: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_during['new_case_sta'] = df_during.new_case.diff()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Make stationary with differencing\n",
    "\n",
    "fig, axes = plt.subplots(4, 1, sharex=True)\n",
    "axes[0].plot(df_during.new_case); axes[0].set_title('Original New Cases')\n",
    "\n",
    "# 1st Differencing\n",
    "axes[1].plot(df_during.new_case.diff()); axes[1].set_title('1st Order Differencing')\n",
    "\n",
    "# 2nd Differencing\n",
    "axes[2].plot(df_during.new_case.diff().diff()); axes[2].set_title('2nd Order Differencing')\n",
    "\n",
    "# 3rd Differencing\n",
    "axes[3].plot(df_during.new_case.diff().diff().diff()); axes[3].set_title('3rd Order Differencing')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "result = adfuller(df_during.new_case.diff().dropna())\n",
    "result\n",
    "\n",
    "df_during['new_case_sta'] = df_during.new_case.diff()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 825,
   "id": "4a858d8d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/2792412098.py:17: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_during['caution_sta'] = df_during.caution.diff()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(4, 1, sharex=True)\n",
    "axes[0].plot(df_during.caution); axes[0].set_title('Original Caution')\n",
    "\n",
    "# 1st Differencing\n",
    "axes[1].plot(df_during.caution.diff()); axes[1].set_title('1st Order Differencing')\n",
    "\n",
    "# 2nd Differencing\n",
    "axes[2].plot(df_during.caution.diff().diff()); axes[2].set_title('2nd Order Differencing')\n",
    "\n",
    "# 3rd Differencing\n",
    "axes[3].plot(df_during.caution.diff().diff().diff()); axes[3].set_title('3rd Order Differencing')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "result = adfuller(df_during.caution.diff().dropna())\n",
    "result\n",
    "\n",
    "df_during['caution_sta'] = df_during.caution.diff()\n",
    "df_during2 = df_during.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 826,
   "id": "42def0ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 1\n",
      "ssr based F test:         F=0.1221  , p=0.7277  , df_denom=81, df_num=1\n",
      "ssr based chi2 test:   chi2=0.1266  , p=0.7220  , df=1\n",
      "likelihood ratio test: chi2=0.1265  , p=0.7221  , df=1\n",
      "parameter F test:         F=0.1221  , p=0.7277  , df_denom=81, df_num=1\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 2\n",
      "ssr based F test:         F=0.5527  , p=0.5777  , df_denom=78, df_num=2\n",
      "ssr based chi2 test:   chi2=1.1762  , p=0.5554  , df=2\n",
      "likelihood ratio test: chi2=1.1679  , p=0.5577  , df=2\n",
      "parameter F test:         F=0.5527  , p=0.5777  , df_denom=78, df_num=2\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 3\n",
      "ssr based F test:         F=3.1558  , p=0.0296  , df_denom=75, df_num=3\n",
      "ssr based chi2 test:   chi2=10.3510 , p=0.0158  , df=3\n",
      "likelihood ratio test: chi2=9.7479  , p=0.0208  , df=3\n",
      "parameter F test:         F=3.1558  , p=0.0296  , df_denom=75, df_num=3\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 4\n",
      "ssr based F test:         F=1.7880  , p=0.1406  , df_denom=72, df_num=4\n",
      "ssr based chi2 test:   chi2=8.0458  , p=0.0899  , df=4\n",
      "likelihood ratio test: chi2=7.6708  , p=0.1044  , df=4\n",
      "parameter F test:         F=1.7880  , p=0.1406  , df_denom=72, df_num=4\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 5\n",
      "ssr based F test:         F=1.3266  , p=0.2633  , df_denom=69, df_num=5\n",
      "ssr based chi2 test:   chi2=7.6902  , p=0.1742  , df=5\n",
      "likelihood ratio test: chi2=7.3427  , p=0.1964  , df=5\n",
      "parameter F test:         F=1.3266  , p=0.2633  , df_denom=69, df_num=5\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 6\n",
      "ssr based F test:         F=2.4104  , p=0.0362  , df_denom=66, df_num=6\n",
      "ssr based chi2 test:   chi2=17.3112 , p=0.0082  , df=6\n",
      "likelihood ratio test: chi2=15.6528 , p=0.0157  , df=6\n",
      "parameter F test:         F=2.4104  , p=0.0362  , df_denom=66, df_num=6\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 7\n",
      "ssr based F test:         F=1.8591  , p=0.0915  , df_denom=63, df_num=7\n",
      "ssr based chi2 test:   chi2=16.1125 , p=0.0241  , df=7\n",
      "likelihood ratio test: chi2=14.6470 , p=0.0408  , df=7\n",
      "parameter F test:         F=1.8591  , p=0.0915  , df_denom=63, df_num=7\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 8\n",
      "ssr based F test:         F=1.7915  , p=0.0965  , df_denom=60, df_num=8\n",
      "ssr based chi2 test:   chi2=18.3931 , p=0.0185  , df=8\n",
      "likelihood ratio test: chi2=16.4934 , p=0.0358  , df=8\n",
      "parameter F test:         F=1.7915  , p=0.0965  , df_denom=60, df_num=8\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 9\n",
      "ssr based F test:         F=1.1097  , p=0.3710  , df_denom=57, df_num=9\n",
      "ssr based chi2 test:   chi2=13.3159 , p=0.1488  , df=9\n",
      "likelihood ratio test: chi2=12.2699 , p=0.1985  , df=9\n",
      "parameter F test:         F=1.1097  , p=0.3710  , df_denom=57, df_num=9\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 10\n",
      "ssr based F test:         F=0.9909  , p=0.4626  , df_denom=54, df_num=10\n",
      "ssr based chi2 test:   chi2=13.7631 , p=0.1841  , df=10\n",
      "likelihood ratio test: chi2=12.6363 , p=0.2447  , df=10\n",
      "parameter F test:         F=0.9909  , p=0.4626  , df_denom=54, df_num=10\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 11\n",
      "ssr based F test:         F=1.1959  , p=0.3136  , df_denom=51, df_num=11\n",
      "ssr based chi2 test:   chi2=19.0881 , p=0.0595  , df=11\n",
      "likelihood ratio test: chi2=16.9816 , p=0.1084  , df=11\n",
      "parameter F test:         F=1.1959  , p=0.3136  , df_denom=51, df_num=11\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 12\n",
      "ssr based F test:         F=1.0022  , p=0.4615  , df_denom=48, df_num=12\n",
      "ssr based chi2 test:   chi2=18.2904 , p=0.1072  , df=12\n",
      "likelihood ratio test: chi2=16.3218 , p=0.1769  , df=12\n",
      "parameter F test:         F=1.0022  , p=0.4615  , df_denom=48, df_num=12\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{1: ({'ssr_ftest': (0.12207036316582402, 0.7277064739121691, 81.0, 1),\n",
       "   'ssr_chi2test': (0.1265914877275212, 0.7219926143225155, 1),\n",
       "   'lrtest': (0.12649619414241897, 0.7220929313345865, 1),\n",
       "   'params_ftest': (0.122070363165888, 0.727706473912092, 81.0, 1.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700157940>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700414e20>,\n",
       "   array([[0., 1., 0.]])]),\n",
       " 2: ({'ssr_ftest': (0.5526594773941723, 0.5776539565890505, 78.0, 2),\n",
       "   'ssr_chi2test': (1.1761727339414436, 0.5553890793542089, 2),\n",
       "   'lrtest': (1.1679170074716012, 0.5576863877413327, 2),\n",
       "   'params_ftest': (0.5526594773941753, 0.5776539565890505, 78.0, 2.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700cb3af0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2770094dd30>,\n",
       "   array([[0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 1., 0.]])]),\n",
       " 3: ({'ssr_ftest': (3.1557791239991535, 0.029634780153914124, 75.0, 3),\n",
       "   'ssr_chi2test': (10.350955526717225, 0.015806797911764647, 3),\n",
       "   'lrtest': (9.747898102219096, 0.020835253720149466, 3),\n",
       "   'params_ftest': (3.1557791239991437, 0.02963478015391455, 75.0, 3.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c31f940>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c31f580>,\n",
       "   array([[0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 4: ({'ssr_ftest': (1.787952047715129, 0.1406224595552599, 72.0, 4),\n",
       "   'ssr_chi2test': (8.04578421471808, 0.08991538455124361, 4),\n",
       "   'lrtest': (7.6708224853750835, 0.10440840887110694, 4),\n",
       "   'params_ftest': (1.7879520477151236, 0.14062245955526145, 72.0, 4.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c31f6a0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c31f760>,\n",
       "   array([[0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 5: ({'ssr_ftest': (1.326562816487629, 0.2632562358166463, 69.0, 5),\n",
       "   'ssr_chi2test': (7.690219226015241, 0.17415493042435806, 5),\n",
       "   'lrtest': (7.342698650066836, 0.1963752404624596, 5),\n",
       "   'params_ftest': (1.3265628164876295, 0.2632562358166463, 69.0, 5.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700041580>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x27700041220>,\n",
       "   array([[0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 6: ({'ssr_ftest': (2.4104165281338834, 0.036247061716160246, 66.0, 6),\n",
       "   'ssr_chi2test': (17.311173247506982, 0.008205164914028463, 6),\n",
       "   'lrtest': (15.652782316743583, 0.015744070680048614, 6),\n",
       "   'params_ftest': (2.4104165281338847, 0.036247061716160246, 66.0, 6.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777bf0c910>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777bd7d2e0>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 7: ({'ssr_ftest': (1.8591377962618505, 0.09150633349549311, 63.0, 7),\n",
       "   'ssr_chi2test': (16.112527567602704, 0.024107942877659914, 7),\n",
       "   'lrtest': (14.647022629855428, 0.0407998403535667, 7),\n",
       "   'params_ftest': (1.859137796261866, 0.09150633349549012, 63.0, 7.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777bd7d460>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777bdf3a00>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 8: ({'ssr_ftest': (1.7915335418046283, 0.09652797728961295, 60.0, 8),\n",
       "   'ssr_chi2test': (18.393077695860853, 0.018465056829058933, 8),\n",
       "   'lrtest': (16.493445670845063, 0.03583793464138944, 8),\n",
       "   'params_ftest': (1.7915335418046356, 0.09652797728961138, 60.0, 8.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c31f0d0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2770073c6d0>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0.]])]),\n",
       " 9: ({'ssr_ftest': (1.1096543118638513, 0.37099231605263716, 57.0, 9),\n",
       "   'ssr_chi2test': (13.31585174236622, 0.1488273445496454, 9),\n",
       "   'lrtest': (12.269868864626574, 0.19851942343161516, 9),\n",
       "   'params_ftest': (1.1096543118638529, 0.37099231605263516, 57.0, 9.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2770073cb80>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2770073c7f0>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0.]])]),\n",
       " 10: ({'ssr_ftest': (0.9909467619032809, 0.4626087312157552, 54.0, 10),\n",
       "   'ssr_chi2test': (13.7631494708789, 0.18407179028675327, 10),\n",
       "   'lrtest': (12.636260017223321, 0.2447257481340748, 10),\n",
       "   'params_ftest': (0.9909467619032789, 0.46260873121575663, 54.0, 10.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c313430>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c313c70>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 1., 0.]])]),\n",
       " 11: ({'ssr_ftest': (1.1959383965492647, 0.3136384441063149, 51.0, 11),\n",
       "   'ssr_chi2test': (19.08811479982552, 0.059530043864611804, 11),\n",
       "   'lrtest': (16.981625253726065, 0.10842028381921293, 11),\n",
       "   'params_ftest': (1.1959383965492576, 0.3136384441063203, 51.0, 11.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d723e20>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d7235e0>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 12: ({'ssr_ftest': (1.0022120891358637, 0.4615005613588257, 48.0, 12),\n",
       "   'ssr_chi2test': (18.290370626729512, 0.10715411104805987, 12),\n",
       "   'lrtest': (16.321768605153466, 0.17693893400741403, 12),\n",
       "   'params_ftest': (1.0022120891358608, 0.46150056135882883, 48.0, 12.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d723790>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277001b5370>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 1., 0.]])])}"
      ]
     },
     "execution_count": 826,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grangercausalitytests(df_during2[['caution_sta','new_case_sta']],maxlag=12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 827,
   "id": "23857c46",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 1\n",
      "ssr based F test:         F=14.6385 , p=0.0003  , df_denom=81, df_num=1\n",
      "ssr based chi2 test:   chi2=15.1807 , p=0.0001  , df=1\n",
      "likelihood ratio test: chi2=13.9546 , p=0.0002  , df=1\n",
      "parameter F test:         F=14.6385 , p=0.0003  , df_denom=81, df_num=1\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 2\n",
      "ssr based F test:         F=7.2379  , p=0.0013  , df_denom=78, df_num=2\n",
      "ssr based chi2 test:   chi2=15.4038 , p=0.0005  , df=2\n",
      "likelihood ratio test: chi2=14.1298 , p=0.0009  , df=2\n",
      "parameter F test:         F=7.2379  , p=0.0013  , df_denom=78, df_num=2\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 3\n",
      "ssr based F test:         F=6.3392  , p=0.0007  , df_denom=75, df_num=3\n",
      "ssr based chi2 test:   chi2=20.7925 , p=0.0001  , df=3\n",
      "likelihood ratio test: chi2=18.5314 , p=0.0003  , df=3\n",
      "parameter F test:         F=6.3392  , p=0.0007  , df_denom=75, df_num=3\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 4\n",
      "ssr based F test:         F=5.2521  , p=0.0009  , df_denom=72, df_num=4\n",
      "ssr based chi2 test:   chi2=23.6344 , p=0.0001  , df=4\n",
      "likelihood ratio test: chi2=20.7379 , p=0.0004  , df=4\n",
      "parameter F test:         F=5.2521  , p=0.0009  , df_denom=72, df_num=4\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 5\n",
      "ssr based F test:         F=4.4528  , p=0.0014  , df_denom=69, df_num=5\n",
      "ssr based chi2 test:   chi2=25.8135 , p=0.0001  , df=5\n",
      "likelihood ratio test: chi2=22.3721 , p=0.0004  , df=5\n",
      "parameter F test:         F=4.4528  , p=0.0014  , df_denom=69, df_num=5\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 6\n",
      "ssr based F test:         F=4.0048  , p=0.0018  , df_denom=66, df_num=6\n",
      "ssr based chi2 test:   chi2=28.7619 , p=0.0001  , df=6\n",
      "likelihood ratio test: chi2=24.5276 , p=0.0004  , df=6\n",
      "parameter F test:         F=4.0048  , p=0.0018  , df_denom=66, df_num=6\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 7\n",
      "ssr based F test:         F=3.2863  , p=0.0048  , df_denom=63, df_num=7\n",
      "ssr based chi2 test:   chi2=28.4814 , p=0.0002  , df=7\n",
      "likelihood ratio test: chi2=24.2784 , p=0.0010  , df=7\n",
      "parameter F test:         F=3.2863  , p=0.0048  , df_denom=63, df_num=7\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 8\n",
      "ssr based F test:         F=2.8017  , p=0.0105  , df_denom=60, df_num=8\n",
      "ssr based chi2 test:   chi2=28.7643 , p=0.0003  , df=8\n",
      "likelihood ratio test: chi2=24.4404 , p=0.0019  , df=8\n",
      "parameter F test:         F=2.8017  , p=0.0105  , df_denom=60, df_num=8\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 9\n",
      "ssr based F test:         F=2.9073  , p=0.0067  , df_denom=57, df_num=9\n",
      "ssr based chi2 test:   chi2=34.8881 , p=0.0001  , df=9\n",
      "likelihood ratio test: chi2=28.7119 , p=0.0007  , df=9\n",
      "parameter F test:         F=2.9073  , p=0.0067  , df_denom=57, df_num=9\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 10\n",
      "ssr based F test:         F=2.4721  , p=0.0162  , df_denom=54, df_num=10\n",
      "ssr based chi2 test:   chi2=34.3353 , p=0.0002  , df=10\n",
      "likelihood ratio test: chi2=28.2698 , p=0.0016  , df=10\n",
      "parameter F test:         F=2.4721  , p=0.0162  , df_denom=54, df_num=10\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 11\n",
      "ssr based F test:         F=3.4430  , p=0.0012  , df_denom=51, df_num=11\n",
      "ssr based chi2 test:   chi2=54.9523 , p=0.0000  , df=11\n",
      "likelihood ratio test: chi2=41.0979 , p=0.0000  , df=11\n",
      "parameter F test:         F=3.4430  , p=0.0012  , df_denom=51, df_num=11\n",
      "\n",
      "Granger Causality\n",
      "number of lags (no zero) 12\n",
      "ssr based F test:         F=3.1392  , p=0.0023  , df_denom=48, df_num=12\n",
      "ssr based chi2 test:   chi2=57.2895 , p=0.0000  , df=12\n",
      "likelihood ratio test: chi2=42.2889 , p=0.0000  , df=12\n",
      "parameter F test:         F=3.1392  , p=0.0023  , df_denom=48, df_num=12\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{1: ({'ssr_ftest': (14.638490669879177, 0.00025505950884945413, 81.0, 1),\n",
       "   'ssr_chi2test': (15.180656990985812, 9.769908858853156e-05, 1),\n",
       "   'lrtest': (13.954601339638373, 0.00018727870304518197, 1),\n",
       "   'params_ftest': (14.638490669879184, 0.0002550595088494528, 81.0, 1.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d070>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777be97be0>,\n",
       "   array([[0., 1., 0.]])]),\n",
       " 2: ({'ssr_ftest': (7.237921927029389, 0.0013079335737837378, 78.0, 2),\n",
       "   'ssr_chi2test': (15.40378256265229, 0.00045197156865048647, 2),\n",
       "   'lrtest': (14.129806804612144, 0.0008545774665873749, 2),\n",
       "   'params_ftest': (7.237921927029397, 0.001307933573783731, 78.0, 2.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277002ebb20>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277002eb9d0>,\n",
       "   array([[0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 1., 0.]])]),\n",
       " 3: ({'ssr_ftest': (6.339166030130282, 0.0006870379173784978, 75.0, 3),\n",
       "   'ssr_chi2test': (20.79246457882732, 0.00011625745862495847, 3),\n",
       "   'lrtest': (18.53140970723348, 0.0003416876178827801, 3),\n",
       "   'params_ftest': (6.339166030130289, 0.0006870379173784952, 75.0, 3.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777bfca910>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86db80>,\n",
       "   array([[0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 4: ({'ssr_ftest': (5.252098189438307, 0.0009072296123517527, 72.0, 4),\n",
       "   'ssr_chi2test': (23.63444185247238, 9.454531877445214e-05, 4),\n",
       "   'lrtest': (20.737912796565695, 0.00035689466337987, 4),\n",
       "   'params_ftest': (5.252098189438304, 0.0009072296123517567, 72.0, 4.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d9d0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d160>,\n",
       "   array([[0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 5: ({'ssr_ftest': (4.4528246171138814, 0.0014211651776671447, 69.0, 5),\n",
       "   'ssr_chi2test': (25.81347604123989, 9.69839454561758e-05, 5),\n",
       "   'lrtest': (22.37209995437115, 0.0004448231093309952, 5),\n",
       "   'params_ftest': (4.4528246171138814, 0.0014211651776671447, 69.0, 5.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d7f0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d760>,\n",
       "   array([[0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 6: ({'ssr_ftest': (4.004817437601582, 0.0017673656377816536, 66.0, 6),\n",
       "   'ssr_chi2test': (28.761870688229543, 6.748488873323119e-05, 6),\n",
       "   'lrtest': (24.527607100668888, 0.00041750872251012803, 6),\n",
       "   'params_ftest': (4.004817437601577, 0.0017673656377816692, 66.0, 6.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d100>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86dd30>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 7: ({'ssr_ftest': (3.2863150442712397, 0.004845118230538526, 63.0, 7),\n",
       "   'ssr_chi2test': (28.481397050350743, 0.00017982261144388614, 7),\n",
       "   'lrtest': (24.278394474264132, 0.0010178051521646718, 7),\n",
       "   'params_ftest': (3.2863150442712294, 0.0048451182305386175, 63.0, 7.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777c081dc0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277002ec3a0>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 8: ({'ssr_ftest': (2.80171458436195, 0.010492654517437252, 60.0, 8),\n",
       "   'ssr_chi2test': (28.764269732782683, 0.0003487580375701076, 8),\n",
       "   'lrtest': (24.44036405124217, 0.001932699386411844, 8),\n",
       "   'params_ftest': (2.8017145843619495, 0.010492654517437252, 60.0, 8.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x2777d86d1f0>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9250>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0.]])]),\n",
       " 9: ({'ssr_ftest': (2.9073409107111927, 0.006674527568894757, 57.0, 9),\n",
       "   'ssr_chi2test': (34.88809092853431, 6.235297862392062e-05, 9),\n",
       "   'lrtest': (28.71190035970983, 0.000724654873783771, 9),\n",
       "   'params_ftest': (2.907340910711199, 0.006674527568894661, 57.0, 9.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9970>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c95b0>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0.]])]),\n",
       " 10: ({'ssr_ftest': (2.472138401628763, 0.016218089534678185, 54.0, 10),\n",
       "   'ssr_chi2test': (34.33525557817726, 0.00016203296380185074, 10),\n",
       "   'lrtest': (28.269809064966694, 0.0016340809963527984, 10),\n",
       "   'params_ftest': (2.472138401628766, 0.016218089534678085, 54.0, 10.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9f40>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9550>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 1., 0.]])]),\n",
       " 11: ({'ssr_ftest': (3.4429566233796964, 0.001247377704467944, 51.0, 11),\n",
       "   'ssr_chi2test': (54.95228806727593, 7.907282693432419e-08, 11),\n",
       "   'lrtest': (41.09792632711469, 2.3182944415730694e-05, 11),\n",
       "   'params_ftest': (3.442956623379699, 0.001247377704467935, 51.0, 11.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9160>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9b80>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 1., 0.]])]),\n",
       " 12: ({'ssr_ftest': (3.139150525503821, 0.0023470301982649764, 48.0, 12),\n",
       "   'ssr_chi2test': (57.289497090444726, 7.007622021097274e-08, 12),\n",
       "   'lrtest': (42.28885869182284, 2.9775954178691763e-05, 12),\n",
       "   'params_ftest': (3.13915052550383, 0.0023470301982649183, 48.0, 12.0)},\n",
       "  [<statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9100>,\n",
       "   <statsmodels.regression.linear_model.RegressionResultsWrapper at 0x277006c9850>,\n",
       "   array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
       "          [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "           0., 0., 0., 0., 0., 0., 0., 1., 0.]])])}"
      ]
     },
     "execution_count": 827,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grangercausalitytests(df_during2[['new_case_sta','caution_sta']],maxlag=12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 828,
   "id": "52a1e96d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(68, 2) (17, 2)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\anaconda3\\lib\\site-packages\\pmdarima\\arima\\_auto_solvers.py:103: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
      "  gen = random_state.permutation(list(gen))[:n_fits]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ARIMA(1,1,2)(0,0,0)[0] intercept   : AIC=-516.328, Time=0.48 sec\n",
      " ARIMA(2,1,2)(0,0,0)[0] intercept   : AIC=-515.392, Time=0.45 sec\n",
      " ARIMA(5,1,0)(0,0,0)[0] intercept   : AIC=-508.038, Time=0.66 sec\n",
      " ARIMA(0,1,1)(0,0,0)[0] intercept   : AIC=-501.087, Time=0.16 sec\n",
      " ARIMA(2,1,0)(0,0,0)[0] intercept   : AIC=-502.055, Time=0.36 sec\n",
      " ARIMA(1,1,4)(0,0,0)[0] intercept   : AIC=-519.759, Time=0.58 sec\n",
      " ARIMA(2,1,3)(0,0,0)[0] intercept   : AIC=-520.904, Time=0.39 sec\n",
      " ARIMA(4,1,0)(0,0,0)[0] intercept   : AIC=-509.614, Time=0.52 sec\n",
      " ARIMA(1,1,0)(0,0,0)[0] intercept   : AIC=-501.816, Time=0.09 sec\n",
      " ARIMA(4,1,1)(0,0,0)[0] intercept   : AIC=-507.766, Time=0.35 sec\n",
      " ARIMA(0,1,4)(0,0,0)[0] intercept   : AIC=-517.191, Time=0.26 sec\n",
      " ARIMA(0,1,2)(0,0,0)[0] intercept   : AIC=-513.767, Time=0.38 sec\n",
      " ARIMA(0,1,5)(0,0,0)[0] intercept   : AIC=-521.035, Time=0.43 sec\n",
      " ARIMA(3,1,1)(0,0,0)[0] intercept   : AIC=-509.406, Time=0.35 sec\n",
      " ARIMA(1,1,3)(0,0,0)[0] intercept   : AIC=-523.833, Time=0.32 sec\n",
      " ARIMA(1,1,1)(0,0,0)[0] intercept   : AIC=-500.241, Time=0.41 sec\n",
      " ARIMA(3,1,2)(0,0,0)[0] intercept   : AIC=-507.551, Time=0.37 sec\n",
      " ARIMA(0,1,3)(0,0,0)[0] intercept   : AIC=-518.439, Time=0.43 sec\n",
      " ARIMA(0,1,0)(0,0,0)[0] intercept   : AIC=-501.397, Time=0.04 sec\n",
      " ARIMA(3,1,0)(0,0,0)[0] intercept   : AIC=-511.544, Time=0.31 sec\n",
      " ARIMA(2,1,1)(0,0,0)[0] intercept   : AIC=-502.258, Time=0.29 sec\n",
      "\n",
      "Best model:  ARIMA(1,1,3)(0,0,0)[0] intercept\n",
      "Total fit time: 7.646 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>68</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>SARIMAX(1, 1, 3)</td> <th>  Log Likelihood     </th>  <td>212.288</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Sat, 11 Dec 2021</td> <th>  AIC                </th> <td>-410.575</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>19:23:56</td>     <th>  BIC                </th> <td>-395.143</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                  <td>0</td>        <th>  HQIC               </th> <td>-404.469</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                       <td> - 68</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>intercept</th> <td>    0.0001</td> <td>    0.000</td> <td>    0.355</td> <td> 0.723</td> <td>   -0.000</td> <td>    0.001</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>        <td>   -0.0071</td> <td>    0.015</td> <td>   -0.467</td> <td> 0.641</td> <td>   -0.037</td> <td>    0.023</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L1</th>     <td>    0.5696</td> <td>    0.176</td> <td>    3.244</td> <td> 0.001</td> <td>    0.225</td> <td>    0.914</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>     <td>   -0.3406</td> <td>    0.143</td> <td>   -2.382</td> <td> 0.017</td> <td>   -0.621</td> <td>   -0.060</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L2</th>     <td>    0.2312</td> <td>    0.145</td> <td>    1.593</td> <td> 0.111</td> <td>   -0.053</td> <td>    0.516</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L3</th>     <td>   -0.7311</td> <td>    0.089</td> <td>   -8.218</td> <td> 0.000</td> <td>   -0.905</td> <td>   -0.557</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>    <td>    0.0001</td> <td> 1.62e-05</td> <td>    6.180</td> <td> 0.000</td> <td> 6.84e-05</td> <td>    0.000</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (L1) (Q):</th>     <td>0.00</td> <th>  Jarque-Bera (JB):  </th> <td>1.42</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.95</td> <th>  Prob(JB):          </th> <td>0.49</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>0.70</td> <th>  Skew:              </th> <td>0.27</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.40</td> <th>  Kurtosis:          </th> <td>3.47</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:                   68\n",
       "Model:               SARIMAX(1, 1, 3)   Log Likelihood                 212.288\n",
       "Date:                Sat, 11 Dec 2021   AIC                           -410.575\n",
       "Time:                        19:23:56   BIC                           -395.143\n",
       "Sample:                             0   HQIC                          -404.469\n",
       "                                 - 68                                         \n",
       "Covariance Type:                  opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "intercept      0.0001      0.000      0.355      0.723      -0.000       0.001\n",
       "x1            -0.0071      0.015     -0.467      0.641      -0.037       0.023\n",
       "ar.L1          0.5696      0.176      3.244      0.001       0.225       0.914\n",
       "ma.L1         -0.3406      0.143     -2.382      0.017      -0.621      -0.060\n",
       "ma.L2          0.2312      0.145      1.593      0.111      -0.053       0.516\n",
       "ma.L3         -0.7311      0.089     -8.218      0.000      -0.905      -0.557\n",
       "sigma2         0.0001   1.62e-05      6.180      0.000    6.84e-05       0.000\n",
       "===================================================================================\n",
       "Ljung-Box (L1) (Q):                   0.00   Jarque-Bera (JB):                 1.42\n",
       "Prob(Q):                              0.95   Prob(JB):                         0.49\n",
       "Heteroskedasticity (H):               0.70   Skew:                             0.27\n",
       "Prob(H) (two-sided):                  0.40   Kurtosis:                         3.47\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 828,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_during2 = df_during.dropna()     # creating this helps indices stay in sync\n",
    "df_sta = df_during2[['new_case_sta','caution_sta']].dropna()\n",
    "train, test = df_sta[0:round(0.8*(len(df_sta.caution_sta)))], df_sta[round(0.8*(len(df_sta.caution_sta))):]\n",
    "print(train.shape, test.shape)\n",
    "\n",
    "model = auto_arima(df_sta.caution_sta, start_p=2, d=1, start_q=1, exogenous=np.reshape(df_sta.new_case_sta.values,(-1,1)),\n",
    "                   trace=1, seasonal= False, stepwise=False, random=True, n_fits=30, random_state=0) \n",
    "fit_sta = model.fit(train.caution_sta, exogenous=np.reshape(train.new_case_sta.values,(-1,1)))\n",
    "\n",
    "fit_sta.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 829,
   "id": "baf98cbc",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ping\\AppData\\Local\\Temp/ipykernel_1420/835576867.py:16: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  test['caution_fc']=df_fc['caution_fc']\n"
     ]
    }
   ],
   "source": [
    "lag_order = 1  # based on model fit\n",
    "forecast_input = train.caution_sta.values[-lag_order:]\n",
    "fc = fit_sta.predict(exogenous=np.reshape(test.new_case_sta.values,(-1,1)), n_periods=len(test))\n",
    "df_forecast = pd.DataFrame(fc,index=df_sta.index[-len(test):], columns = ['caution_sta_fc'])\n",
    "\n",
    "df_fc = df_forecast.copy()\n",
    "df_check = pd.DataFrame()\n",
    "\n",
    "# De-difference\n",
    "df_fc['caution_fc'] = df_during2['caution'].iloc[len(train)-1] + df_fc['caution_sta_fc'].cumsum()\n",
    "\n",
    "# Check   \n",
    "df_check['caution_check'] = df_during2['caution'].iloc[len(train)-1] + test[\"caution_sta\"].cumsum()\n",
    "df_check['caution'] = df_during2['caution'].iloc[len(train):]\n",
    "\n",
    "test['caution_fc']=df_fc['caution_fc']\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 830,
   "id": "38bdc8c5",
   "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>caution_check</th>\n",
       "      <th>caution</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2021-08-08</th>\n",
       "      <td>0.316195</td>\n",
       "      <td>0.316195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-08-15</th>\n",
       "      <td>0.321419</td>\n",
       "      <td>0.321419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-08-22</th>\n",
       "      <td>0.329469</td>\n",
       "      <td>0.329469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-08-29</th>\n",
       "      <td>0.334845</td>\n",
       "      <td>0.334845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-09-05</th>\n",
       "      <td>0.335321</td>\n",
       "      <td>0.335321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-09-12</th>\n",
       "      <td>0.339387</td>\n",
       "      <td>0.339387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-09-19</th>\n",
       "      <td>0.342114</td>\n",
       "      <td>0.342114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-09-26</th>\n",
       "      <td>0.344588</td>\n",
       "      <td>0.344588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-10-03</th>\n",
       "      <td>0.343433</td>\n",
       "      <td>0.343433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-10-10</th>\n",
       "      <td>0.340573</td>\n",
       "      <td>0.340573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-10-17</th>\n",
       "      <td>0.344099</td>\n",
       "      <td>0.344099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-10-24</th>\n",
       "      <td>0.335884</td>\n",
       "      <td>0.335884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-10-31</th>\n",
       "      <td>0.331168</td>\n",
       "      <td>0.331168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-11-07</th>\n",
       "      <td>0.326046</td>\n",
       "      <td>0.326046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-11-14</th>\n",
       "      <td>0.320517</td>\n",
       "      <td>0.320517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-11-21</th>\n",
       "      <td>0.314581</td>\n",
       "      <td>0.314581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-11-28</th>\n",
       "      <td>0.308240</td>\n",
       "      <td>0.308240</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            caution_check   caution\n",
       "2021-08-08       0.316195  0.316195\n",
       "2021-08-15       0.321419  0.321419\n",
       "2021-08-22       0.329469  0.329469\n",
       "2021-08-29       0.334845  0.334845\n",
       "2021-09-05       0.335321  0.335321\n",
       "2021-09-12       0.339387  0.339387\n",
       "2021-09-19       0.342114  0.342114\n",
       "2021-09-26       0.344588  0.344588\n",
       "2021-10-03       0.343433  0.343433\n",
       "2021-10-10       0.340573  0.340573\n",
       "2021-10-17       0.344099  0.344099\n",
       "2021-10-24       0.335884  0.335884\n",
       "2021-10-31       0.331168  0.331168\n",
       "2021-11-07       0.326046  0.326046\n",
       "2021-11-14       0.320517  0.320517\n",
       "2021-11-21       0.314581  0.314581\n",
       "2021-11-28       0.308240  0.308240"
      ]
     },
     "execution_count": 830,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_check"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 831,
   "id": "4909a774",
   "metadata": {},
   "outputs": [
    {
     "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": [
    "fig,ax = plt.subplots()\n",
    "\n",
    "ax.plot(df_during2.caution[round(0.8*(len(df_sta.caution_sta))):],color='black')\n",
    "ax.set_ylabel('New cases per day',color='black')\n",
    "ax.set_xlabel('Time')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "ax.plot(test.caution_fc,color='red',linestyle='--')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "edd6dd63",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}