Add grangers_causation_matrix()

granger_causality_testing.ipynb

@@ -2,20 +2,23 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [],
    "source": [
+    "import numpy as np\n",
     "import pandas as pd\n",
     "import plotly.express as px\n",
+    "from statsmodels.stats.stattools import durbin_watson\n",
     "from statsmodels.tsa.api import VAR\n",
     "from statsmodels.tsa.stattools import adfuller\n",
+    "from statsmodels.tsa.stattools import grangercausalitytests\n",
     "from statsmodels.tsa.stattools import kpss"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -6934,6 +6937,372 @@
     "    print('HQIC: ', result.hqic, '\\n')"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "  Summary of Regression Results   \n",
+       "==================================\n",
+       "Model:                         VAR\n",
+       "Method:                        OLS\n",
+       "Date:           Sun, 25, Feb, 2024\n",
+       "Time:                     21:40:09\n",
+       "--------------------------------------------------------------------\n",
+       "No. of Equations:         3.00000    BIC:                    14.5979\n",
+       "Nobs:                     128.000    HQIC:                   12.7724\n",
+       "Log likelihood:          -1144.35    FPE:                    111728.\n",
+       "AIC:                      11.5231    Det(Omega_mle):         44477.8\n",
+       "--------------------------------------------------------------------\n",
+       "Results for equation apple\n",
+       "==============================================================================\n",
+       "                 coefficient       std. error           t-stat            prob\n",
+       "------------------------------------------------------------------------------\n",
+       "const              -0.699238         0.584786           -1.196           0.232\n",
+       "L1.apple            0.014847         0.140038            0.106           0.916\n",
+       "L1.walmart         -0.113138         0.105005           -1.077           0.281\n",
+       "L1.tesla            0.081087         0.038983            2.080           0.038\n",
+       "L2.apple           -0.167780         0.135627           -1.237           0.216\n",
+       "L2.walmart         -0.031901         0.104304           -0.306           0.760\n",
+       "L2.tesla            0.109001         0.039080            2.789           0.005\n",
+       "L3.apple            0.085712         0.136509            0.628           0.530\n",
+       "L3.walmart          0.157405         0.102108            1.542           0.123\n",
+       "L3.tesla           -0.081028         0.038433           -2.108           0.035\n",
+       "L4.apple           -0.440097         0.112933           -3.897           0.000\n",
+       "L4.walmart         -0.028011         0.102538           -0.273           0.785\n",
+       "L4.tesla            0.196295         0.038369            5.116           0.000\n",
+       "L5.apple            0.187708         0.125176            1.500           0.134\n",
+       "L5.walmart         -0.059267         0.102669           -0.577           0.564\n",
+       "L5.tesla           -0.089774         0.042168           -2.129           0.033\n",
+       "L6.apple           -0.021143         0.124388           -0.170           0.865\n",
+       "L6.walmart         -0.236903         0.105059           -2.255           0.024\n",
+       "L6.tesla            0.021333         0.040789            0.523           0.601\n",
+       "L7.apple            0.034935         0.121061            0.289           0.773\n",
+       "L7.walmart         -0.196441         0.105540           -1.861           0.063\n",
+       "L7.tesla            0.124142         0.039365            3.154           0.002\n",
+       "L8.apple            0.263352         0.114911            2.292           0.022\n",
+       "L8.walmart          0.124926         0.105791            1.181           0.238\n",
+       "L8.tesla           -0.064692         0.037117           -1.743           0.081\n",
+       "L9.apple           -0.161093         0.118119           -1.364           0.173\n",
+       "L9.walmart          0.142378         0.105329            1.352           0.176\n",
+       "L9.tesla           -0.003729         0.036768           -0.101           0.919\n",
+       "L10.apple           0.052834         0.118872            0.444           0.657\n",
+       "L10.walmart        -0.012512         0.106188           -0.118           0.906\n",
+       "L10.tesla           0.014551         0.036445            0.399           0.690\n",
+       "L11.apple          -0.196153         0.120203           -1.632           0.103\n",
+       "L11.walmart        -0.189079         0.103137           -1.833           0.067\n",
+       "L11.tesla           0.046769         0.033933            1.378           0.168\n",
+       "L12.apple           0.281041         0.113450            2.477           0.013\n",
+       "L12.walmart         0.248444         0.097517            2.548           0.011\n",
+       "L12.tesla          -0.131892         0.034494           -3.824           0.000\n",
+       "L13.apple          -0.033669         0.107860           -0.312           0.755\n",
+       "L13.walmart         0.034415         0.101916            0.338           0.736\n",
+       "L13.tesla          -0.018055         0.036536           -0.494           0.621\n",
+       "L14.apple          -0.255695         0.103146           -2.479           0.013\n",
+       "L14.walmart         0.139620         0.095459            1.463           0.144\n",
+       "L14.tesla           0.100374         0.036446            2.754           0.006\n",
+       "L15.apple           0.196757         0.104793            1.878           0.060\n",
+       "L15.walmart         0.017978         0.097782            0.184           0.854\n",
+       "L15.tesla          -0.030502         0.036230           -0.842           0.400\n",
+       "==============================================================================\n",
+       "\n",
+       "Results for equation walmart\n",
+       "==============================================================================\n",
+       "                 coefficient       std. error           t-stat            prob\n",
+       "------------------------------------------------------------------------------\n",
+       "const              -0.103331         0.633837           -0.163           0.870\n",
+       "L1.apple            0.144375         0.151784            0.951           0.342\n",
+       "L1.walmart         -0.076127         0.113813           -0.669           0.504\n",
+       "L1.tesla           -0.048764         0.042252           -1.154           0.248\n",
+       "L2.apple           -0.125694         0.147003           -0.855           0.393\n",
+       "L2.walmart         -0.017047         0.113053           -0.151           0.880\n",
+       "L2.tesla           -0.008949         0.042358           -0.211           0.833\n",
+       "L3.apple            0.162286         0.147959            1.097           0.273\n",
+       "L3.walmart         -0.060079         0.110672           -0.543           0.587\n",
+       "L3.tesla            0.031460         0.041656            0.755           0.450\n",
+       "L4.apple            0.080359         0.122405            0.657           0.512\n",
+       "L4.walmart         -0.068210         0.111139           -0.614           0.539\n",
+       "L4.tesla            0.023108         0.041587            0.556           0.578\n",
+       "L5.apple            0.025574         0.135675            0.188           0.850\n",
+       "L5.walmart         -0.232336         0.111281           -2.088           0.037\n",
+       "L5.tesla            0.020971         0.045705            0.459           0.646\n",
+       "L6.apple           -0.201452         0.134822           -1.494           0.135\n",
+       "L6.walmart         -0.134846         0.113871           -1.184           0.236\n",
+       "L6.tesla            0.068019         0.044210            1.539           0.124\n",
+       "L7.apple            0.186968         0.131215            1.425           0.154\n",
+       "L7.walmart         -0.110714         0.114393           -0.968           0.333\n",
+       "L7.tesla           -0.016329         0.042667           -0.383           0.702\n",
+       "L8.apple            0.175006         0.124550            1.405           0.160\n",
+       "L8.walmart         -0.016260         0.114665           -0.142           0.887\n",
+       "L8.tesla           -0.072481         0.040230           -1.802           0.072\n",
+       "L9.apple            0.198956         0.128027            1.554           0.120\n",
+       "L9.walmart         -0.084669         0.114164           -0.742           0.458\n",
+       "L9.tesla           -0.004090         0.039852           -0.103           0.918\n",
+       "L10.apple          -0.091041         0.128843           -0.707           0.480\n",
+       "L10.walmart         0.052878         0.115095            0.459           0.646\n",
+       "L10.tesla          -0.003499         0.039502           -0.089           0.929\n",
+       "L11.apple          -0.212327         0.130285           -1.630           0.103\n",
+       "L11.walmart        -0.065408         0.111788           -0.585           0.558\n",
+       "L11.tesla           0.070514         0.036780            1.917           0.055\n",
+       "L12.apple           0.341238         0.122966            2.775           0.006\n",
+       "L12.walmart        -0.128880         0.105697           -1.219           0.223\n",
+       "L12.tesla          -0.035130         0.037387           -0.940           0.347\n",
+       "L13.apple          -0.009063         0.116907           -0.078           0.938\n",
+       "L13.walmart         0.138496         0.110464            1.254           0.210\n",
+       "L13.tesla          -0.016126         0.039601           -0.407           0.684\n",
+       "L14.apple           0.050583         0.111798            0.452           0.651\n",
+       "L14.walmart         0.123218         0.103466            1.191           0.234\n",
+       "L14.tesla           0.005768         0.039503            0.146           0.884\n",
+       "L15.apple           0.153644         0.113583            1.353           0.176\n",
+       "L15.walmart        -0.039947         0.105984           -0.377           0.706\n",
+       "L15.tesla          -0.054155         0.039269           -1.379           0.168\n",
+       "==============================================================================\n",
+       "\n",
+       "Results for equation tesla\n",
+       "==============================================================================\n",
+       "                 coefficient       std. error           t-stat            prob\n",
+       "------------------------------------------------------------------------------\n",
+       "const              -2.063049         1.878392           -1.098           0.272\n",
+       "L1.apple           -0.171895         0.449815           -0.382           0.702\n",
+       "L1.walmart         -0.105253         0.337288           -0.312           0.755\n",
+       "L1.tesla            0.190132         0.125216            1.518           0.129\n",
+       "L2.apple           -0.046830         0.435648           -0.107           0.914\n",
+       "L2.walmart         -0.125800         0.335034           -0.375           0.707\n",
+       "L2.tesla            0.315252         0.125529            2.511           0.012\n",
+       "L3.apple            0.806135         0.438482            1.838           0.066\n",
+       "L3.walmart         -0.279177         0.327980           -0.851           0.395\n",
+       "L3.tesla           -0.426441         0.123450           -3.454           0.001\n",
+       "L4.apple           -0.685840         0.362751           -1.891           0.059\n",
+       "L4.walmart          0.055336         0.329363            0.168           0.867\n",
+       "L4.tesla            0.397299         0.123244            3.224           0.001\n",
+       "L5.apple            0.071231         0.402077            0.177           0.859\n",
+       "L5.walmart         -0.108462         0.329783           -0.329           0.742\n",
+       "L5.tesla           -0.050922         0.135447           -0.376           0.707\n",
+       "L6.apple           -0.389905         0.399548           -0.976           0.329\n",
+       "L6.walmart         -0.362135         0.337460           -1.073           0.283\n",
+       "L6.tesla           -0.029577         0.131017           -0.226           0.821\n",
+       "L7.apple            0.402571         0.388859            1.035           0.301\n",
+       "L7.walmart         -0.773885         0.339006           -2.283           0.022\n",
+       "L7.tesla            0.070900         0.126445            0.561           0.575\n",
+       "L8.apple            0.057018         0.369107            0.154           0.877\n",
+       "L8.walmart         -0.253383         0.339812           -0.746           0.456\n",
+       "L8.tesla           -0.060940         0.119222           -0.511           0.609\n",
+       "L9.apple           -0.814436         0.379410           -2.147           0.032\n",
+       "L9.walmart          0.692883         0.338329            2.048           0.041\n",
+       "L9.tesla            0.170148         0.118103            1.441           0.150\n",
+       "L10.apple           0.165284         0.381830            0.433           0.665\n",
+       "L10.walmart        -0.070246         0.341086           -0.206           0.837\n",
+       "L10.tesla           0.111133         0.117065            0.949           0.342\n",
+       "L11.apple          -0.257494         0.386104           -0.667           0.505\n",
+       "L11.walmart        -0.834862         0.331288           -2.520           0.012\n",
+       "L11.tesla           0.318169         0.108998            2.919           0.004\n",
+       "L12.apple           1.048785         0.364413            2.878           0.004\n",
+       "L12.walmart         0.390809         0.313235            1.248           0.212\n",
+       "L12.tesla          -0.268612         0.110798           -2.424           0.015\n",
+       "L13.apple          -0.201846         0.346456           -0.583           0.560\n",
+       "L13.walmart         0.385814         0.327363            1.179           0.239\n",
+       "L13.tesla          -0.250162         0.117359           -2.132           0.033\n",
+       "L14.apple          -0.934654         0.331317           -2.821           0.005\n",
+       "L14.walmart        -0.212581         0.306624           -0.693           0.488\n",
+       "L14.tesla           0.296885         0.117067            2.536           0.011\n",
+       "L15.apple           1.015922         0.336607            3.018           0.003\n",
+       "L15.walmart        -0.438696         0.314087           -1.397           0.162\n",
+       "L15.tesla          -0.205261         0.116373           -1.764           0.078\n",
+       "==============================================================================\n",
+       "\n",
+       "Correlation matrix of residuals\n",
+       "              apple   walmart     tesla\n",
+       "apple      1.000000  0.281952  0.608528\n",
+       "walmart    0.281952  1.000000  0.187816\n",
+       "tesla      0.608528  0.187816  1.000000\n",
+       "\n"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "results = model.fit(maxlags=15, ic='aic')\n",
+    "results.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "out = durbin_watson(results.resid)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "apple : 2.12\n",
+      "walmart : 1.9\n",
+      "tesla : 2.13\n"
+     ]
+    }
+   ],
+   "source": [
+    "for col, val in zip(df.columns, out):\n",
+    "    print(col, ':', round(val, 2))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "maxlag=15\n",
+    "test = 'ssr_chi2test'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def grangers_causation_matrix(data, variables, test='ssr_chi2test', verbose=False):    \n",
+    "   \n",
+    "    df = pd.DataFrame(np.zeros((len(variables), len(variables))), columns=variables, index=variables)\n",
+    "    for c in df.columns:\n",
+    "        for r in df.index:\n",
+    "            test_result = grangercausalitytests(data[[r, c]], maxlag=maxlag, verbose=False)\n",
+    "            p_values = [round(test_result[i+1][0][test][1],4) for i in range(maxlag)]\n",
+    "            if verbose: print(f'Y = {r}, X = {c}, P Values = {p_values}')\n",
+    "            min_p_value = np.min(p_values)\n",
+    "            df.loc[r, c] = min_p_value\n",
+    "    df.columns = [var + '_x' for var in variables]\n",
+    "    df.index = [var + '_y' for var in variables]\n",
+    "    return df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\n",
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/statsmodels/tsa/stattools.py:1545: FutureWarning:\n",
+      "\n",
+      "verbose is deprecated since functions should not print results\n",
+      "\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>apple_x</th>\n",
+       "      <th>walmart_x</th>\n",
+       "      <th>tesla_x</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>apple_y</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0003</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>walmart_y</th>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1.0000</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>tesla_y</th>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0000</td>\n",
+       "      <td>1.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "           apple_x  walmart_x  tesla_x\n",
+       "apple_y        1.0     0.0003      0.0\n",
+       "walmart_y      0.0     1.0000      0.0\n",
+       "tesla_y        0.0     0.0000      1.0"
+      ]
+     },
+     "execution_count": 38,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "grangers_causation_matrix(df_train_transformed, variables = df_train_transformed.columns)"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,