Add grangers_causation_matrix()

granger_causality_testing.ipynb

@@ -2,13 +2,15 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 43,
    "metadata": {},
    "outputs": [],
    "source": [
     "import numpy as np\n",
     "import pandas as pd\n",
     "import plotly.express as px\n",
+    "from sklearn.metrics import mean_absolute_error\n",
+    "from sklearn.metrics import mean_squared_error\n",
     "from statsmodels.stats.stattools import durbin_watson\n",
     "from statsmodels.tsa.api import VAR\n",
     "from statsmodels.tsa.stattools import adfuller\n",
@@ -7303,6 +7305,78 @@
     "grangers_causation_matrix(df_train_transformed, variables = df_train_transformed.columns)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "lag_order = results.k_ar\n",
+    "\n",
+    "df_input = df_train_transformed.values[-lag_order:]\n",
+    "df_forecast = results.forecast(y=df_input, steps=n_obs)\n",
+    "df_forecast = (pd.DataFrame(df_forecast, index=df_test.index, columns=df_test.columns + '_pred'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def invert_transformation(df, pred):\n",
+    "    forecast = df_forecast.copy()\n",
+    "    columns = df.columns\n",
+    "    for col in columns:\n",
+    "        forecast[str(col)+'_pred'] = df[col].iloc[-1] + forecast[str(col)+'_pred'].cumsum()\n",
+    "    return forecast"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "output = invert_transformation(df_train, df_forecast)\n",
+    "\n",
+    "combined = pd.concat([output['apple_pred'], df_test['apple'], output['walmart_pred'], df_test['walmart'], output['tesla_pred'], df_test['tesla']], axis=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Forecast accuracy of Apple\n",
+      "RMSE:  4.37\n",
+      "MAE:  3.67\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/ibnu/miniconda3/envs/py312/lib/python3.12/site-packages/sklearn/metrics/_regression.py:483: FutureWarning:\n",
+      "\n",
+      "'squared' is deprecated in version 1.4 and will be removed in 1.6. To calculate the root mean squared error, use the function'root_mean_squared_error'.\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "rmse = mean_squared_error(combined['apple_pred'], combined['apple'], squared=False)\n",
+    "mae = mean_absolute_error(combined['apple_pred'], combined['apple'])\n",
+    "\n",
+    "print('Forecast accuracy of Apple')\n",
+    "print('RMSE: ', round(rmse,2))\n",
+    "print('MAE: ', round(mae,2))"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,