{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "5df158e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the packages\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import copy\n",
    "from numpy.linalg import inv\n",
    "from pandas_datareader.data import DataReader\n",
    "from pandas.plotting import register_matplotlib_converters\n",
    "register_matplotlib_converters()\n",
    "\n",
    "from scipy import signal\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.tsa.api import VAR\n",
    "from scipy.optimize import root\n",
    "from statsmodels.tsa.vector_ar.vecm import coint_johansen\n",
    "from statsmodels.compat import lzip"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f42180f",
   "metadata": {},
   "source": [
    "# An example with simulated data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "9668a38f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate some sample data\n",
    "np.random.seed(0)\n",
    "n_obs = 100\n",
    "data = np.random.normal(size=(n_obs, 2))\n",
    "data += np.arange(n_obs)[:, np.newaxis]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07f1c5be",
   "metadata": {},
   "source": [
    "Pandas DataFrame is a widely-used data structure in Python for handling structured data, such as time series data commonly used in VAR modeling. Its intuitive interface, labeling and indexing capabilities, rich functionality, and seamless integration with other libraries make it the preferred choice for data manipulation and VAR estimation in Python."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "3dc65973",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a DataFrame\n",
    "df = pd.DataFrame(data, columns=['X1', 'X2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "2dabd39d",
   "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": [
    "# Plot the time series data\n",
    "df.plot()\n",
    "plt.title('Sample Time Series Data')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Value')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "738bc715",
   "metadata": {},
   "source": [
    "In Python, fitting a VAR model involves using the VAR class from the statsmodels library. This module provides comprehensive functionality for time series analysis, including VAR modeling. With statsmodels, we can easily specify the order of the VAR model, which determines the number of lagged observations included in the model. This flexibility allows us to tailor the model complexity to the characteristics of the time series data and the underlying system dynamics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "5079a90e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9941657275847339"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adfuller(df['X1'])[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "3af31b34",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['dX1'] = df['X1'] -  df[\"X1\"].shift(1)\n",
    "df['dX2'] = df['X2'] -  df[\"X2\"].shift(1)\n",
    "df = df[1:] #eliminate NaN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "6b778d96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-8.357729339498562,\n",
       " 2.878705776525002e-13,\n",
       " 4,\n",
       " 94,\n",
       " {'1%': -3.5019123847798657,\n",
       "  '5%': -2.892815255482889,\n",
       "  '10%': -2.583453861475781},\n",
       " 247.65394031411506)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adfuller(df['dX1'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "91dc23f7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function adfuller in module statsmodels.tsa.stattools:\n",
      "\n",
      "adfuller(x, maxlag=None, regression='c', autolag='AIC', store=False, regresults=False)\n",
      "    Augmented Dickey-Fuller unit root test.\n",
      "    \n",
      "    The Augmented Dickey-Fuller test can be used to test for a unit root in a\n",
      "    univariate process in the presence of serial correlation.\n",
      "    \n",
      "    Parameters\n",
      "    ----------\n",
      "    x : array_like, 1d\n",
      "        The data series to test.\n",
      "    maxlag : int\n",
      "        Maximum lag which is included in test, default 12*(nobs/100)^{1/4}.\n",
      "    regression : {\"c\",\"ct\",\"ctt\",\"nc\"}\n",
      "        Constant and trend order to include in regression.\n",
      "    \n",
      "        * \"c\" : constant only (default).\n",
      "        * \"ct\" : constant and trend.\n",
      "        * \"ctt\" : constant, and linear and quadratic trend.\n",
      "        * \"nc\" : no constant, no trend.\n",
      "    \n",
      "    autolag : {\"AIC\", \"BIC\", \"t-stat\", None}\n",
      "        Method to use when automatically determining the lag length among the\n",
      "        values 0, 1, ..., maxlag.\n",
      "    \n",
      "        * If \"AIC\" (default) or \"BIC\", then the number of lags is chosen\n",
      "          to minimize the corresponding information criterion.\n",
      "        * \"t-stat\" based choice of maxlag.  Starts with maxlag and drops a\n",
      "          lag until the t-statistic on the last lag length is significant\n",
      "          using a 5%-sized test.\n",
      "        * If None, then the number of included lags is set to maxlag.\n",
      "    store : bool\n",
      "        If True, then a result instance is returned additionally to\n",
      "        the adf statistic. Default is False.\n",
      "    regresults : bool, optional\n",
      "        If True, the full regression results are returned. Default is False.\n",
      "    \n",
      "    Returns\n",
      "    -------\n",
      "    adf : float\n",
      "        The test statistic.\n",
      "    pvalue : float\n",
      "        MacKinnon\"s approximate p-value based on MacKinnon (1994, 2010).\n",
      "    usedlag : int\n",
      "        The number of lags used.\n",
      "    nobs : int\n",
      "        The number of observations used for the ADF regression and calculation\n",
      "        of the critical values.\n",
      "    critical values : dict\n",
      "        Critical values for the test statistic at the 1 %, 5 %, and 10 %\n",
      "        levels. Based on MacKinnon (2010).\n",
      "    icbest : float\n",
      "        The maximized information criterion if autolag is not None.\n",
      "    resstore : ResultStore, optional\n",
      "        A dummy class with results attached as attributes.\n",
      "    \n",
      "    Notes\n",
      "    -----\n",
      "    The null hypothesis of the Augmented Dickey-Fuller is that there is a unit\n",
      "    root, with the alternative that there is no unit root. If the pvalue is\n",
      "    above a critical size, then we cannot reject that there is a unit root.\n",
      "    \n",
      "    The p-values are obtained through regression surface approximation from\n",
      "    MacKinnon 1994, but using the updated 2010 tables. If the p-value is close\n",
      "    to significant, then the critical values should be used to judge whether\n",
      "    to reject the null.\n",
      "    \n",
      "    The autolag option and maxlag for it are described in Greene.\n",
      "    \n",
      "    References\n",
      "    ----------\n",
      "    .. [1] W. Green.  \"Econometric Analysis,\" 5th ed., Pearson, 2003.\n",
      "    \n",
      "    .. [2] Hamilton, J.D.  \"Time Series Analysis\".  Princeton, 1994.\n",
      "    \n",
      "    .. [3] MacKinnon, J.G. 1994.  \"Approximate asymptotic distribution functions for\n",
      "        unit-root and cointegration tests.  `Journal of Business and Economic\n",
      "        Statistics` 12, 167-76.\n",
      "    \n",
      "    .. [4] MacKinnon, J.G. 2010. \"Critical Values for Cointegration Tests.\"  Queen\"s\n",
      "        University, Dept of Economics, Working Papers.  Available at\n",
      "        http://ideas.repec.org/p/qed/wpaper/1227.html\n",
      "    \n",
      "    Examples\n",
      "    --------\n",
      "    See example notebook\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(adfuller)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "777d3217",
   "metadata": {
    "scrolled": false
   },
   "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>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>dX1</th>\n",
       "      <th>dX2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.978738</td>\n",
       "      <td>3.240893</td>\n",
       "      <td>0.214686</td>\n",
       "      <td>2.840736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.867558</td>\n",
       "      <td>1.022722</td>\n",
       "      <td>1.888820</td>\n",
       "      <td>-2.218171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.950088</td>\n",
       "      <td>2.848643</td>\n",
       "      <td>0.082530</td>\n",
       "      <td>1.825921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.896781</td>\n",
       "      <td>4.410599</td>\n",
       "      <td>-0.053307</td>\n",
       "      <td>1.561956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5.144044</td>\n",
       "      <td>6.454274</td>\n",
       "      <td>1.247262</td>\n",
       "      <td>2.043675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>93.707143</td>\n",
       "      <td>95.267051</td>\n",
       "      <td>0.446706</td>\n",
       "      <td>-0.275964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>95.960717</td>\n",
       "      <td>94.831907</td>\n",
       "      <td>2.253574</td>\n",
       "      <td>-0.435144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>97.523277</td>\n",
       "      <td>96.828454</td>\n",
       "      <td>1.562559</td>\n",
       "      <td>1.996547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>98.771791</td>\n",
       "      <td>98.823504</td>\n",
       "      <td>1.248514</td>\n",
       "      <td>1.995050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>101.163236</td>\n",
       "      <td>100.336528</td>\n",
       "      <td>2.391445</td>\n",
       "      <td>1.513024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>99 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            X1          X2       dX1       dX2\n",
       "1     1.978738    3.240893  0.214686  2.840736\n",
       "2     3.867558    1.022722  1.888820 -2.218171\n",
       "3     3.950088    2.848643  0.082530  1.825921\n",
       "4     3.896781    4.410599 -0.053307  1.561956\n",
       "5     5.144044    6.454274  1.247262  2.043675\n",
       "..         ...         ...       ...       ...\n",
       "95   93.707143   95.267051  0.446706 -0.275964\n",
       "96   95.960717   94.831907  2.253574 -0.435144\n",
       "97   97.523277   96.828454  1.562559  1.996547\n",
       "98   98.771791   98.823504  1.248514  1.995050\n",
       "99  101.163236  100.336528  2.391445  1.513024\n",
       "\n",
       "[99 rows x 4 columns]"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "32c792f7",
   "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>dX1</th>\n",
       "      <th>dX2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.214686</td>\n",
       "      <td>2.840736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.888820</td>\n",
       "      <td>-2.218171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.082530</td>\n",
       "      <td>1.825921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.053307</td>\n",
       "      <td>1.561956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.247262</td>\n",
       "      <td>2.043675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>0.446706</td>\n",
       "      <td>-0.275964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>2.253574</td>\n",
       "      <td>-0.435144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>1.562559</td>\n",
       "      <td>1.996547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>1.248514</td>\n",
       "      <td>1.995050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>2.391445</td>\n",
       "      <td>1.513024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>99 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         dX1       dX2\n",
       "1   0.214686  2.840736\n",
       "2   1.888820 -2.218171\n",
       "3   0.082530  1.825921\n",
       "4  -0.053307  1.561956\n",
       "5   1.247262  2.043675\n",
       "..       ...       ...\n",
       "95  0.446706 -0.275964\n",
       "96  2.253574 -0.435144\n",
       "97  1.562559  1.996547\n",
       "98  1.248514  1.995050\n",
       "99  2.391445  1.513024\n",
       "\n",
       "[99 rows x 2 columns]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.loc[:,[\"dX1\",\"dX2\"]].copy() # we only want dX1 and dX2\n",
    "# df.loc: Access a group of rows and columns by label(s)\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "775164c0",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Summary of Regression Results   \n",
      "==================================\n",
      "Model:                         VAR\n",
      "Method:                        OLS\n",
      "Date:           Thu, 29, Feb, 2024\n",
      "Time:                     08:37:51\n",
      "--------------------------------------------------------------------\n",
      "No. of Equations:         2.00000    BIC:                   0.891520\n",
      "Nobs:                     95.0000    HQIC:                  0.603156\n",
      "Log likelihood:          -270.961    FPE:                    1.50496\n",
      "AIC:                     0.407627    Det(Omega_mle):         1.25576\n",
      "--------------------------------------------------------------------\n",
      "Results for equation dX1\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          3.030040         0.542375            5.587           0.000\n",
      "L1.dX1        -0.525980         0.105346           -4.993           0.000\n",
      "L1.dX2         0.061463         0.101769            0.604           0.546\n",
      "L2.dX1        -0.782845         0.114944           -6.811           0.000\n",
      "L2.dX2        -0.080062         0.122655           -0.653           0.514\n",
      "L3.dX1        -0.360150         0.115447           -3.120           0.002\n",
      "L3.dX2        -0.018219         0.121665           -0.150           0.881\n",
      "L4.dX1        -0.324838         0.107887           -3.011           0.003\n",
      "L4.dX2        -0.002540         0.098681           -0.026           0.979\n",
      "=========================================================================\n",
      "\n",
      "Results for equation dX2\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          2.809534         0.561401            5.005           0.000\n",
      "L1.dX1         0.062854         0.109041            0.576           0.564\n",
      "L1.dX2        -0.854404         0.105338           -8.111           0.000\n",
      "L2.dX1         0.080940         0.118976            0.680           0.496\n",
      "L2.dX2        -0.645462         0.126957           -5.084           0.000\n",
      "L3.dX1         0.226850         0.119497            1.898           0.058\n",
      "L3.dX2        -0.515125         0.125932           -4.090           0.000\n",
      "L4.dX1         0.094890         0.111671            0.850           0.395\n",
      "L4.dX2        -0.248172         0.102143           -2.430           0.015\n",
      "=========================================================================\n",
      "\n",
      "Correlation matrix of residuals\n",
      "            dX1       dX2\n",
      "dX1    1.000000 -0.105070\n",
      "dX2   -0.105070  1.000000\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Fit VAR model\n",
    "model = VAR(X)\n",
    "order = 4  # Order of VAR model\n",
    "results = model.fit(order)\n",
    "\n",
    "# Print summary of the model\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54e02736",
   "metadata": {},
   "source": [
    "## Practical Example"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd87a522",
   "metadata": {},
   "source": [
    "The seminal paper titled \"The Dynamic Effects of Aggregate Demand and Supply Disturbances,\" published in the American Economic Review in 1989, Blanchard and Quah propose a methodology to decompose fluctuations in output and prices into components attributable to aggregate demand (AD) and aggregate supply (AS) disturbances. They argue that understanding the relative contributions of demand and supply shocks to economic fluctuations is essential for policymakers to formulate effective stabilization policies.\n",
    "\n",
    "Blanchard and Quah use a vector autoregressive (VAR) framework to analyze the dynamics of output and prices in the United States. The dataset they employ consists of quarterly observations spanning the period from 1953 to 1986."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "c29bba34",
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Downloading and preparing data using DataReader\n",
    "'''\n",
    "           \n",
    "#Sample selection\n",
    "start = '1948Q1'                                              \n",
    "end = '1987Q3'\n",
    "\n",
    "#Data from FRED\n",
    "gdp = DataReader('GNPC96', 'fred', start=start,  end=end) # quarterly frequency\n",
    "urate = DataReader('UNRATE', 'fred', start=start, end=end)  # monthly frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "56d9393c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Frequency change for urate\n",
    "urateq = urate.resample('Q').mean() # Convenience method for frequency conversion and resampling of time series. \n",
    "                                    # Object must have a datetime-like index \n",
    "# urateq_test = urate.groupby(pd.PeriodIndex(urate.index, freq='Q'), axis=0).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "448bcc6d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "Plot original data\n",
    "'''\n",
    "\n",
    "plt.figure(figsize=(16,10))\n",
    "plt.subplot(211)\n",
    "plt.plot(gdp)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Annualized quarterly GDP')\n",
    "plt.title('Annualized quarterly GDP')\n",
    "plt.grid()\n",
    "plt.subplot(212)\n",
    "plt.plot(urateq)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Unemployment rate')\n",
    "plt.title('Unemployment rate')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "2afdacfa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function adfuller in module statsmodels.tsa.stattools:\n",
      "\n",
      "adfuller(x, maxlag=None, regression='c', autolag='AIC', store=False, regresults=False)\n",
      "    Augmented Dickey-Fuller unit root test.\n",
      "    \n",
      "    The Augmented Dickey-Fuller test can be used to test for a unit root in a\n",
      "    univariate process in the presence of serial correlation.\n",
      "    \n",
      "    Parameters\n",
      "    ----------\n",
      "    x : array_like, 1d\n",
      "        The data series to test.\n",
      "    maxlag : int\n",
      "        Maximum lag which is included in test, default 12*(nobs/100)^{1/4}.\n",
      "    regression : {\"c\",\"ct\",\"ctt\",\"nc\"}\n",
      "        Constant and trend order to include in regression.\n",
      "    \n",
      "        * \"c\" : constant only (default).\n",
      "        * \"ct\" : constant and trend.\n",
      "        * \"ctt\" : constant, and linear and quadratic trend.\n",
      "        * \"nc\" : no constant, no trend.\n",
      "    \n",
      "    autolag : {\"AIC\", \"BIC\", \"t-stat\", None}\n",
      "        Method to use when automatically determining the lag length among the\n",
      "        values 0, 1, ..., maxlag.\n",
      "    \n",
      "        * If \"AIC\" (default) or \"BIC\", then the number of lags is chosen\n",
      "          to minimize the corresponding information criterion.\n",
      "        * \"t-stat\" based choice of maxlag.  Starts with maxlag and drops a\n",
      "          lag until the t-statistic on the last lag length is significant\n",
      "          using a 5%-sized test.\n",
      "        * If None, then the number of included lags is set to maxlag.\n",
      "    store : bool\n",
      "        If True, then a result instance is returned additionally to\n",
      "        the adf statistic. Default is False.\n",
      "    regresults : bool, optional\n",
      "        If True, the full regression results are returned. Default is False.\n",
      "    \n",
      "    Returns\n",
      "    -------\n",
      "    adf : float\n",
      "        The test statistic.\n",
      "    pvalue : float\n",
      "        MacKinnon\"s approximate p-value based on MacKinnon (1994, 2010).\n",
      "    usedlag : int\n",
      "        The number of lags used.\n",
      "    nobs : int\n",
      "        The number of observations used for the ADF regression and calculation\n",
      "        of the critical values.\n",
      "    critical values : dict\n",
      "        Critical values for the test statistic at the 1 %, 5 %, and 10 %\n",
      "        levels. Based on MacKinnon (2010).\n",
      "    icbest : float\n",
      "        The maximized information criterion if autolag is not None.\n",
      "    resstore : ResultStore, optional\n",
      "        A dummy class with results attached as attributes.\n",
      "    \n",
      "    Notes\n",
      "    -----\n",
      "    The null hypothesis of the Augmented Dickey-Fuller is that there is a unit\n",
      "    root, with the alternative that there is no unit root. If the pvalue is\n",
      "    above a critical size, then we cannot reject that there is a unit root.\n",
      "    \n",
      "    The p-values are obtained through regression surface approximation from\n",
      "    MacKinnon 1994, but using the updated 2010 tables. If the p-value is close\n",
      "    to significant, then the critical values should be used to judge whether\n",
      "    to reject the null.\n",
      "    \n",
      "    The autolag option and maxlag for it are described in Greene.\n",
      "    \n",
      "    References\n",
      "    ----------\n",
      "    .. [1] W. Green.  \"Econometric Analysis,\" 5th ed., Pearson, 2003.\n",
      "    \n",
      "    .. [2] Hamilton, J.D.  \"Time Series Analysis\".  Princeton, 1994.\n",
      "    \n",
      "    .. [3] MacKinnon, J.G. 1994.  \"Approximate asymptotic distribution functions for\n",
      "        unit-root and cointegration tests.  `Journal of Business and Economic\n",
      "        Statistics` 12, 167-76.\n",
      "    \n",
      "    .. [4] MacKinnon, J.G. 2010. \"Critical Values for Cointegration Tests.\"  Queen\"s\n",
      "        University, Dept of Economics, Working Papers.  Available at\n",
      "        http://ideas.repec.org/p/qed/wpaper/1227.html\n",
      "    \n",
      "    Examples\n",
      "    --------\n",
      "    See example notebook\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(adfuller)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "04b99ebc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "urateq:\n",
      "ADF Statistic: -2.379583651816758\n",
      "p-value: 0.39052768099433177\n",
      "gdp:\n",
      "ADF Statistic: 1.8241666832764825\n",
      "p-value: 0.998395762517703\n"
     ]
    }
   ],
   "source": [
    "# Check stationarity of series\n",
    "from statsmodels.tsa.stattools import adfuller\n",
    "\n",
    "# Assuming 'data' is your time series data\n",
    "result_1 = adfuller(urateq, regression='ct')\n",
    "\n",
    "# Extracting test statistics and p-value\n",
    "adf_statistic = result_1[0]\n",
    "p_value = result_1[1]\n",
    "\n",
    "# Printing test results\n",
    "print('urateq:')\n",
    "print(\"ADF Statistic:\", adf_statistic)\n",
    "print(\"p-value:\", p_value)\n",
    "\n",
    "# Assuming 'data' is your time series data\n",
    "result_2 = adfuller(gdp)\n",
    "\n",
    "# Extracting test statistics and p-value\n",
    "adf_statistic = result_2[0]\n",
    "p_value = result_2[1]\n",
    "\n",
    "# Printing test results\n",
    "print('gdp:')\n",
    "print(\"ADF Statistic:\", adf_statistic)\n",
    "print(\"p-value:\", p_value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "8244d002",
   "metadata": {},
   "outputs": [],
   "source": [
    "#From YYYYQ format to date format (find it important for matplotlib)\n",
    "gdp.index = pd.date_range(start,'1987Q4',freq=\"Q\")\n",
    "urateq.index = pd.date_range(start,'1987Q4',freq=\"Q\")\n",
    "\n",
    "'''\n",
    "Transforming the data part 2\n",
    "'''\n",
    "\n",
    "\n",
    "#create Data Frame\n",
    "df = pd.concat((urateq, gdp), axis=1)\n",
    "df.columns = [\"urate\",\"gdp\"]\n",
    "\n",
    "\n",
    "# Differenciate GDP and urate\n",
    "df[\"lngdp\"] = np.log(df[\"gdp\"])\n",
    "df[\"dy\"] = (df[\"lngdp\"] - df[\"lngdp\"].shift(1))*400 #annualized quarterly GDP growth\n",
    "df[\"durates\"] = (df[\"urate\"] - df[\"urate\"].shift(1))\n",
    "df = df[1:] # drop first observation\n",
    "\n",
    "\n",
    "# Detrending as in the paper: Important to make the data covariance stationary. \n",
    "# Weak stationarity is an important property for representation and estimation\n",
    "# Productivity Slowdown less apparent than in origal data, upward trend in urate present\n",
    "# signal.detrend: Remove linear trend along axis from data.\n",
    "\n",
    "df.loc[:\"1973Q2\",\"dys\"] = signal.detrend(df.loc[:\"1973Q2\",\"dy\"],type=\"constant\")\n",
    "df.loc[\"1973Q3\":,\"dys\"] = signal.detrend(df.loc[\"1973Q3\":,\"dy\"],type=\"constant\")\n",
    "df[\"urates\"] = signal.detrend(df[\"urate\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "42ddcbc6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "'''\n",
    "Plot new data\n",
    "'''\n",
    "\n",
    "plt.figure(figsize=(16,10))\n",
    "plt.subplot(211)\n",
    "plt.plot(df['dy'],label='observed')\n",
    "plt.plot(df['dys'],label='detrended')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Annualized quarterly GDP growth')\n",
    "plt.title('Annualized quarterly GDP growth')\n",
    "plt.grid()\n",
    "plt.subplot(212)\n",
    "plt.plot(df['urate'],label='observed')\n",
    "plt.plot(df['durates'],label='detrended')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Unemployment rate')\n",
    "plt.title('Unemployment rate')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "8ae12af7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pandas.core.frame.DataFrame"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Setting up the data for VAR\n",
    "X = df.loc[:,[\"dys\",\"durates\"]].copy()\n",
    "type(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a049358f",
   "metadata": {},
   "source": [
    "Python is an object oriented programming language. In Python, the dynamic nature of typing means we must be cautious about the type of objects we work with. Objects can change type during runtime, leading to unexpected behavior if operations are not appropriate for the object's type. Being mindful of object types helps prevent errors and ensures code reliability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "7745fdd0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>VAR Order Selection (* highlights the minimums)</caption>\n",
       "<tr>\n",
       "   <td></td>      <th>AIC</th>         <th>BIC</th>         <th>FPE</th>        <th>HQIC</th>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>0</th>  <td>    0.4222</td>  <td>    0.4222</td>  <td>     1.525</td>  <td>    0.4222</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>1</th>  <td>   -0.1181</td>  <td>  -0.03631*</td> <td>    0.8887</td>  <td>  -0.08484</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>2</th>  <td>   -0.1988</td>  <td>  -0.03536</td>  <td>    0.8197</td>  <td>   -0.1324*</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>3</th>  <td>   -0.2150*</td> <td>   0.03018</td>  <td>    0.8066*</td> <td>   -0.1154</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>4</th>  <td>   -0.1893</td>  <td>    0.1377</td>  <td>    0.8277</td>  <td>  -0.05644</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>5</th>  <td>   -0.1503</td>  <td>    0.2585</td>  <td>    0.8609</td>  <td>   0.01581</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>6</th>  <td>   -0.1522</td>  <td>    0.3382</td>  <td>    0.8594</td>  <td>   0.04706</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>7</th>  <td>   -0.1096</td>  <td>    0.4626</td>  <td>    0.8972</td>  <td>    0.1229</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>8</th>  <td>  -0.09753</td>  <td>    0.5564</td>  <td>    0.9087</td>  <td>    0.1682</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>9</th>  <td>  -0.08049</td>  <td>    0.6552</td>  <td>    0.9250</td>  <td>    0.2184</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>10</th> <td>  -0.05495</td>  <td>    0.7625</td>  <td>    0.9498</td>  <td>    0.2772</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>11</th> <td>  -0.07041</td>  <td>    0.8288</td>  <td>    0.9363</td>  <td>    0.2949</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>12</th> <td>   -0.1040</td>  <td>    0.8769</td>  <td>    0.9066</td>  <td>    0.2945</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.table.SimpleTable'>"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = VAR(X)\n",
    "model.select_order(12, trend='n').summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "e43e54f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "  Summary of Regression Results   \n",
       "==================================\n",
       "Model:                         VAR\n",
       "Method:                        OLS\n",
       "Date:           Thu, 29, Feb, 2024\n",
       "Time:                     08:49:49\n",
       "--------------------------------------------------------------------\n",
       "No. of Equations:         2.00000    BIC:                   0.136324\n",
       "Nobs:                     155.000    HQIC:               -0.00359260\n",
       "Log likelihood:          -420.175    FPE:                   0.905545\n",
       "AIC:                   -0.0992961    Det(Omega_mle):        0.839308\n",
       "--------------------------------------------------------------------\n",
       "Results for equation dys\n",
       "=============================================================================\n",
       "                coefficient       std. error           t-stat            prob\n",
       "-----------------------------------------------------------------------------\n",
       "L1.dys             0.082186         0.111721            0.736           0.462\n",
       "L1.durates        -4.213621         1.409532           -2.989           0.003\n",
       "L2.dys             0.171330         0.117187            1.462           0.144\n",
       "L2.durates         2.936790         1.427493            2.057           0.040\n",
       "L3.dys            -0.014437         0.118171           -0.122           0.903\n",
       "L3.durates         1.278716         1.196988            1.068           0.285\n",
       "=============================================================================\n",
       "\n",
       "Results for equation durates\n",
       "=============================================================================\n",
       "                coefficient       std. error           t-stat            prob\n",
       "-----------------------------------------------------------------------------\n",
       "L1.dys            -0.033853         0.008772           -3.859           0.000\n",
       "L1.durates         0.360218         0.110672            3.255           0.001\n",
       "L2.dys            -0.035749         0.009201           -3.885           0.000\n",
       "L2.durates        -0.353222         0.112082           -3.151           0.002\n",
       "L3.dys            -0.018766         0.009278           -2.023           0.043\n",
       "L3.durates        -0.202550         0.093984           -2.155           0.031\n",
       "=============================================================================\n",
       "\n",
       "Correlation matrix of residuals\n",
       "                dys   durates\n",
       "dys        1.000000 -0.696561\n",
       "durates   -0.696561  1.000000\n",
       "\n"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = VAR(X)\n",
    "results = model.fit(3, trend='n')\n",
    "results.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "94660da1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The VAR model is stable.\n"
     ]
    }
   ],
   "source": [
    "# Stability Condition: roots are larger than one in absolute value.\n",
    "results = model.fit(3, trend='n')\n",
    "roots = results.roots\n",
    "all_stable = all(np.abs(root) > 1 for root in roots)\n",
    "if all_stable:\n",
    "    print(\"The VAR model is stable.\")\n",
    "else:\n",
    "    print(\"The VAR model is not stable.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fd006fc",
   "metadata": {},
   "source": [
    "The ML estimator of the VAR parameters is equivalent to equation by equation OLS estimators.\n",
    "\n",
    "We can write the model in a bigger one and use OLS which is equivalent to Maximum Likelihood for an unrestricted VAR\n",
    "$$X = X_L\\hat{B} + EPS$$\n",
    "where $X$ is $T\\times n$,$X_L$ is $T\\times (n\\times p)$ and $\\hat{B}$ is $(n\\times p) \\times n$\n",
    "\n",
    "$\\hat{B}$ = $[\\hat{B}(1)...\\hat{B}(p)]$\n",
    "\n",
    "Sometimes it is written as as $$X'=\\hat{B}'X_L'+EPS'$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "aeeca6fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "VAR with a OLS\n",
    "'''\n",
    "XLAG = pd.DataFrame() #empty Dataframe\n",
    "\n",
    "num_lags = 3\n",
    "for i in range(1,num_lags+1):\n",
    "    XLAG = pd.concat([XLAG,X.shift(i).add_suffix(\"-\"+str(i))],axis=1)\n",
    "    \n",
    "#change names to frames that we modify    \n",
    "X2 = X.iloc[num_lags:,:]\n",
    "XLAG2 = XLAG.iloc[num_lags:,:]\n",
    "num_vars = X2.shape[1]\n",
    "num_obs = X2.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f25fe84b",
   "metadata": {},
   "source": [
    "The resulting ML estimator of the VAR model parameters is:\n",
    "\\begin{equation}\n",
    "    \\hat{B} = \\bigg[\\sum^T_{t=1} x_t x^´_t \\bigg]^{-1}\\bigg[\\sum^T_{t=1} y_t x^´_t \\bigg]\n",
    "\\end{equation}\n",
    "\n",
    "In matrix form:\n",
    "\n",
    "\\begin{equation}\n",
    "    \\hat{B} = (X^´X)^{-1} X^\\prime Y\n",
    "\\end{equation}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "f73c06d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pandas.core.frame.DataFrame"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# But what is this?\n",
    "type(X2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb9087a1",
   "metadata": {},
   "source": [
    "We want to invert and use algebra, so we need to transform this into arrays: $\\textbf{numpy package}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "116239f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#changing to arrays \n",
    "X3 = np.array(X2)\n",
    "XLAG3 = np.array(XLAG2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "afbb64b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#VAR - standard OLS\n",
    "Bhat = inv(XLAG3.T@XLAG3)@XLAG3.T@X3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "469cb6da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[ 0.08218595, -4.21362131],\n",
       "        [-0.03385338,  0.36021834]],\n",
       "\n",
       "       [[ 0.17133029,  2.93679032],\n",
       "        [-0.03574939, -0.35322194]],\n",
       "\n",
       "       [[-0.0144375 ,  1.27871604],\n",
       "        [-0.01876582, -0.20254968]]])"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(results.coefs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "0bb3d544",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.08218595, -0.03385338],\n",
       "       [-4.21362131,  0.36021834],\n",
       "       [ 0.17133029, -0.03574939],\n",
       "       [ 2.93679032, -0.35322194],\n",
       "       [-0.0144375 , -0.01876582],\n",
       "       [ 1.27871604, -0.20254968]])"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Bhat"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a6d7dc4",
   "metadata": {},
   "source": [
    "# Curse of Dimensionality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "bc017734",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Summary of Regression Results   \n",
      "==================================\n",
      "Model:                         VAR\n",
      "Method:                        OLS\n",
      "Date:           Thu, 29, Feb, 2024\n",
      "Time:                     08:39:33\n",
      "--------------------------------------------------------------------\n",
      "No. of Equations:         5.00000    BIC:                    4.62553\n",
      "Nobs:                     95.0000    HQIC:                   2.94341\n",
      "Log likelihood:          -654.630    FPE:                    6.29607\n",
      "AIC:                      1.80283    Det(Omega_mle):         2.31952\n",
      "--------------------------------------------------------------------\n",
      "Results for equation dX1\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          1.890404         0.907272            2.084           0.037\n",
      "L1.dX1        -0.854525         0.112110           -7.622           0.000\n",
      "L1.dX2        -0.029397         0.118141           -0.249           0.803\n",
      "L1.dX3         0.041910         0.099722            0.420           0.674\n",
      "L1.dX4         0.351082         0.113541            3.092           0.002\n",
      "L1.dX5        -0.052940         0.112118           -0.472           0.637\n",
      "L2.dX1        -0.563420         0.148464           -3.795           0.000\n",
      "L2.dX2        -0.218043         0.151511           -1.439           0.150\n",
      "L2.dX3        -0.004402         0.115050           -0.038           0.969\n",
      "L2.dX4         0.198797         0.150001            1.325           0.185\n",
      "L2.dX5         0.021117         0.144744            0.146           0.884\n",
      "L3.dX1        -0.232093         0.149939           -1.548           0.122\n",
      "L3.dX2        -0.004884         0.150619           -0.032           0.974\n",
      "L3.dX3         0.068166         0.114703            0.594           0.552\n",
      "L3.dX4         0.067669         0.148779            0.455           0.649\n",
      "L3.dX5         0.043505         0.149487            0.291           0.771\n",
      "L4.dX1        -0.061571         0.108924           -0.565           0.572\n",
      "L4.dX2         0.194458         0.125145            1.554           0.120\n",
      "L4.dX3        -0.031570         0.100519           -0.314           0.753\n",
      "L4.dX4         0.102338         0.106941            0.957           0.339\n",
      "L4.dX5         0.072428         0.118054            0.614           0.540\n",
      "=========================================================================\n",
      "\n",
      "Results for equation dX2\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          2.547837         0.896477            2.842           0.004\n",
      "L1.dX1        -0.044126         0.110776           -0.398           0.690\n",
      "L1.dX2        -0.761888         0.116736           -6.527           0.000\n",
      "L1.dX3        -0.052477         0.098535           -0.533           0.594\n",
      "L1.dX4        -0.088536         0.112190           -0.789           0.430\n",
      "L1.dX5         0.020987         0.110784            0.189           0.850\n",
      "L2.dX1        -0.134185         0.146698           -0.915           0.360\n",
      "L2.dX2        -0.499004         0.149708           -3.333           0.001\n",
      "L2.dX3        -0.058320         0.113681           -0.513           0.608\n",
      "L2.dX4         0.003448         0.148216            0.023           0.981\n",
      "L2.dX5         0.301905         0.143021            2.111           0.035\n",
      "L3.dX1        -0.235860         0.148155           -1.592           0.111\n",
      "L3.dX2        -0.293062         0.148827           -1.969           0.049\n",
      "L3.dX3         0.113852         0.113339            1.005           0.315\n",
      "L3.dX4         0.035370         0.147009            0.241           0.810\n",
      "L3.dX5         0.150347         0.147709            1.018           0.309\n",
      "L4.dX1        -0.123722         0.107628           -1.150           0.250\n",
      "L4.dX2        -0.005429         0.123656           -0.044           0.965\n",
      "L4.dX3         0.058670         0.099323            0.591           0.555\n",
      "L4.dX4         0.127892         0.105669            1.210           0.226\n",
      "L4.dX5        -0.089835         0.116649           -0.770           0.441\n",
      "=========================================================================\n",
      "\n",
      "Results for equation dX3\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          3.539535         1.009415            3.507           0.000\n",
      "L1.dX1        -0.174705         0.124732           -1.401           0.161\n",
      "L1.dX2        -0.086476         0.131442           -0.658           0.511\n",
      "L1.dX3        -0.709443         0.110949           -6.394           0.000\n",
      "L1.dX4         0.017159         0.126323            0.136           0.892\n",
      "L1.dX5        -0.071147         0.124740           -0.570           0.568\n",
      "L2.dX1        -0.194202         0.165179           -1.176           0.240\n",
      "L2.dX2        -0.006201         0.168568           -0.037           0.971\n",
      "L2.dX3        -0.483580         0.128002           -3.778           0.000\n",
      "L2.dX4         0.201397         0.166888            1.207           0.228\n",
      "L2.dX5        -0.122249         0.161039           -0.759           0.448\n",
      "L3.dX1        -0.183160         0.166820           -1.098           0.272\n",
      "L3.dX2        -0.150422         0.167577           -0.898           0.369\n",
      "L3.dX3        -0.451235         0.127617           -3.536           0.000\n",
      "L3.dX4         0.223093         0.165529            1.348           0.178\n",
      "L3.dX5         0.049087         0.166317            0.295           0.768\n",
      "L4.dX1        -0.132683         0.121187           -1.095           0.274\n",
      "L4.dX2        -0.246121         0.139235           -1.768           0.077\n",
      "L4.dX3        -0.236353         0.111836           -2.113           0.035\n",
      "L4.dX4         0.101067         0.118981            0.849           0.396\n",
      "L4.dX5         0.103228         0.131344            0.786           0.432\n",
      "=========================================================================\n",
      "\n",
      "Results for equation dX4\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          2.912114         0.903040            3.225           0.001\n",
      "L1.dX1         0.000724         0.111587            0.006           0.995\n",
      "L1.dX2         0.431613         0.117590            3.670           0.000\n",
      "L1.dX3         0.000577         0.099256            0.006           0.995\n",
      "L1.dX4        -0.829822         0.113011           -7.343           0.000\n",
      "L1.dX5        -0.071051         0.111595           -0.637           0.524\n",
      "L2.dX1        -0.064537         0.147772           -0.437           0.662\n",
      "L2.dX2         0.255662         0.150804            1.695           0.090\n",
      "L2.dX3        -0.109783         0.114513           -0.959           0.338\n",
      "L2.dX4        -0.543591         0.149301           -3.641           0.000\n",
      "L2.dX5        -0.114235         0.144069           -0.793           0.428\n",
      "L3.dX1        -0.164512         0.149240           -1.102           0.270\n",
      "L3.dX2         0.128069         0.149917            0.854           0.393\n",
      "L3.dX3        -0.081626         0.114168           -0.715           0.475\n",
      "L3.dX4        -0.218561         0.148085           -1.476           0.140\n",
      "L3.dX5        -0.226757         0.148790           -1.524           0.128\n",
      "L4.dX1        -0.130845         0.108416           -1.207           0.227\n",
      "L4.dX2        -0.037430         0.124562           -0.300           0.764\n",
      "L4.dX3        -0.077877         0.100051           -0.778           0.436\n",
      "L4.dX4        -0.014264         0.106442           -0.134           0.893\n",
      "L4.dX5        -0.030886         0.117503           -0.263           0.793\n",
      "=========================================================================\n",
      "\n",
      "Results for equation dX5\n",
      "=========================================================================\n",
      "            coefficient       std. error           t-stat            prob\n",
      "-------------------------------------------------------------------------\n",
      "const          3.654168         0.927164            3.941           0.000\n",
      "L1.dX1        -0.072009         0.114568           -0.629           0.530\n",
      "L1.dX2         0.041315         0.120732            0.342           0.732\n",
      "L1.dX3        -0.044024         0.101908           -0.432           0.666\n",
      "L1.dX4        -0.138417         0.116030           -1.193           0.233\n",
      "L1.dX5        -0.888588         0.114576           -7.755           0.000\n",
      "L2.dX1        -0.134342         0.151719           -0.885           0.376\n",
      "L2.dX2         0.012919         0.154833            0.083           0.934\n",
      "L2.dX3         0.029937         0.117572            0.255           0.799\n",
      "L2.dX4        -0.152226         0.153289           -0.993           0.321\n",
      "L2.dX5        -0.555755         0.147917           -3.757           0.000\n",
      "L3.dX1        -0.065251         0.153227           -0.426           0.670\n",
      "L3.dX2         0.031371         0.153922            0.204           0.838\n",
      "L3.dX3         0.036315         0.117218            0.310           0.757\n",
      "L3.dX4         0.006109         0.152041            0.040           0.968\n",
      "L3.dX5        -0.451053         0.152765           -2.953           0.003\n",
      "L4.dX1         0.019222         0.111312            0.173           0.863\n",
      "L4.dX2        -0.136672         0.127889           -1.069           0.285\n",
      "L4.dX3        -0.063686         0.102723           -0.620           0.535\n",
      "L4.dX4         0.060457         0.109286            0.553           0.580\n",
      "L4.dX5        -0.183036         0.120642           -1.517           0.129\n",
      "=========================================================================\n",
      "\n",
      "Correlation matrix of residuals\n",
      "            dX1       dX2       dX3       dX4       dX5\n",
      "dX1    1.000000 -0.087806 -0.130156  0.030315 -0.095998\n",
      "dX2   -0.087806  1.000000 -0.038216 -0.004736  0.141856\n",
      "dX3   -0.130156 -0.038216  1.000000 -0.013419 -0.034031\n",
      "dX4    0.030315 -0.004736 -0.013419  1.000000  0.048651\n",
      "dX5   -0.095998  0.141856 -0.034031  0.048651  1.000000\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Generate sample data with 5 variables\n",
    "np.random.seed(0)\n",
    "n_obs = 100\n",
    "n_variables = 5  # Number of variables\n",
    "data = np.random.normal(size=(n_obs, n_variables))\n",
    "data += np.arange(n_obs)[:, np.newaxis]\n",
    "\n",
    "# Create a DataFrame\n",
    "df = pd.DataFrame(data, columns=['X1', 'X2', 'X3', 'X4', 'X5'])\n",
    "\n",
    "df['dX1'] = df['X1'] -  df[\"X1\"].shift(1)\n",
    "df['dX2'] = df['X2'] -  df[\"X2\"].shift(1)\n",
    "df['dX3'] = df['X3'] -  df[\"X3\"].shift(1)\n",
    "df['dX4'] = df['X4'] -  df[\"X4\"].shift(1)\n",
    "df['dX5'] = df['X5'] -  df[\"X5\"].shift(1)\n",
    "df = df[1:] #eliminate NaN\n",
    "\n",
    "X = df.loc[:,[\"dX1\",\"dX2\",\"dX3\",\"dX4\",\"dX5\"]].copy()\n",
    "\n",
    "# Fit VAR model\n",
    "model = VAR(X)\n",
    "order = 4  # Order of VAR model\n",
    "results = model.fit(order)\n",
    "\n",
    "# Print summary of the model\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "205e89be",
   "metadata": {},
   "source": [
    "All standard errors are bigger, indicating a less efficient estimation. In the VAR model with more variables, the standard errors tend to increase relative to the VAR model with fewer variables. This increase in standard errors indicates a higher level of uncertainty in the parameter estimates due to the larger number of parameters being estimated simultaneously."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}