VAR.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from statsmodels.tsa.api import VAR\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_path = '/Users/maxim_anisimov/Desktop/GitHub/data/'\n",
    "\n",
    "cpi_path = 'global-economic-monitor/Core CPI, seas. adj..xlsx'\n",
    "cpi = pd.read_excel(data_path+cpi_path, sheet_name='monthly', index_col=0, parse_dates=True)\n",
    "cpi_us = cpi[['United States']]\n",
    "cpi_growth_us = cpi_us.pct_change()\n",
    "\n",
    "# exchange_rate_path = 'global-economic-monitor/Exchange rate, new LCU per USD extended backward, period average.xlsx'\n",
    "# exchange_rate = pd.read_excel(data_path+exchange_rate_path, sheet_name='monthly', index_col=0, parse_dates=True)\n",
    "# exchange_rate_us = exchange_rate[['United States']]\n",
    "\n",
    "# unemployment_path = 'global-economic-monitor/Unemployment Rate, seas. adj..xlsx'\n",
    "# unemployment = pd.read_excel(data_path+unemployment_path, sheet_name='monthly', index_col=0, parse_dates=True)\n",
    "# unemployment_us = unemployment[['United States']]\n",
    "\n",
    "ind_production_path = 'global-economic-monitor/Industrial Production, constant 2010 US$, seas. adj..xlsx'\n",
    "ind_production = pd.read_excel(data_path+ind_production_path, sheet_name='monthly', index_col=0, parse_dates=True)\n",
    "ind_production_us = ind_production[['United States']]\n",
    "ind_prod_growth_us = ind_production_us.pct_change()\n",
    "\n",
    "interest_rate_path = 'DGS1.xls' # from St. Loius FED\n",
    "interest_rate_us = pd.read_excel(data_path+interest_rate_path, skiprows=10, index_col=0, parse_dates=True)\n",
    "interest_rate_us = interest_rate_us.resample('1M').last() # resample from daily to monthly\n",
    "interest_rate_us = interest_rate_us[interest_rate_us.index < '2019-06-01']\n",
    "interest_rate_us.index = ind_prod_growth_us.index[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_df = pd.concat([cpi_growth_us, ind_prod_growth_us, interest_rate_us], axis=1).dropna()\n",
    "data_df.columns = ['CPI growth', 'Industrial production growth', 'Interest rate']\n",
    "n_variables = data_df.shape[1]\n",
    "n_obs = data_df.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "(data_df/data_df.iloc[0,:]).plot(figsize=(14,7));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = np.asmatrix(data_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Simple VAR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[ 1.06007528e-03,  1.52514982e-04,  4.90776363e-02],\n",
       "        [ 8.18738147e-02,  2.64177843e-01, -3.13674354e+00],\n",
       "        [-4.09705267e-03,  7.60173471e-02,  2.77190854e+00],\n",
       "        [ 1.31690824e-04,  6.26504703e-03,  1.31160993e+00],\n",
       "        [ 1.52147938e-01, -1.22091956e-01, -2.02144795e+01],\n",
       "        [ 6.99007077e-04,  1.52766893e-01,  2.21810425e-01],\n",
       "        [-8.27967242e-06, -5.92688710e-03, -3.18219956e-01]])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_lags = 2\n",
    "include_const = True\n",
    "\n",
    "# transform data\n",
    "endog_variables_lst = [data[lag-1:-n_lags+lag-1] for lag in range(1,n_lags+1)][::-1]\n",
    "X = np.hstack(endog_variables_lst)\n",
    "if include_const == True:\n",
    "    X = np.append(np.ones((n_obs-n_lags,1)), X, axis=1)\n",
    "Y = data[n_lags:]\n",
    "\n",
    "# OLS estimation\n",
    "beta_array = np.linalg.inv(X.T @ X) @ X.T @ Y # transpose if you want to premultiply X.T\n",
    "beta_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[1.70634552e-03, 8.28483039e-04, 2.37654944e+00]])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Forecast 1 period ahead\n",
    "Y_hat_1 = np.asmatrix(X[-1]) @ np.asmatrix(beta_array)\n",
    "Y_hat_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[0.  , 0.  , 2.21]])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y[-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[ 0.00000000e+00,  0.00000000e+00,  2.21000000e+00],\n",
       "        [ 1.38099278e-03, -3.38983051e-03,  2.39000000e+00]])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.vstack([Y[-1], Y[-2]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[ 1.53906927e-03, -8.53453233e-04,  2.15852193e+00],\n",
       "        [ 1.45554085e-03,  9.19046652e-04,  2.16975697e+00],\n",
       "        [ 1.67691545e-03,  1.08892639e-03,  2.17474872e+00],\n",
       "        [ 1.68343888e-03,  1.40598473e-03,  2.17957889e+00],\n",
       "        [ 1.71706917e-03,  1.43140957e-03,  2.18074681e+00]])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "horizon = 5\n",
    "\n",
    "# create new X for the first forecast horizon\n",
    "X_new = np.hstack( [Y[-h] for h in range(1,n_lags+1)] ) # # all needed X vectors stacked in a matrix\n",
    "if include_const:\n",
    "    X_new = np.append(1, X_new)\n",
    "\n",
    "X_hat = np.asmatrix(X_new) # since it is a first iteration\n",
    "\n",
    "# last forecast (here it is number one)\n",
    "Y_hat_last = np.asmatrix(X_hat[-1]) @ np.asmatrix(beta_array)\n",
    "Y_hat = np.asmatrix( np.vstack([Y[-n_lags:], Y_hat_last]) )\n",
    "\n",
    "if horizon > 1:\n",
    "    for i in range(1, horizon):\n",
    "\n",
    "        # create new X now using Y_hat\n",
    "        X_new = np.hstack( [Y_hat[-h] for h in range(1,n_lags+1)] ) # # all needed X vectors stacked in a matrix\n",
    "        if include_const:\n",
    "            X_new = np.append(1, X_new)\n",
    "\n",
    "        X_hat = np.vstack([X_hat, X_new]) # add current X to X_hat\n",
    "\n",
    "        # forecast using the last X_hat\n",
    "        Y_hat_last = np.asmatrix(X_hat[-1]) @ np.asmatrix(beta_array)\n",
    "        Y_hat = np.asmatrix( np.vstack([Y_hat, Y_hat_last]) )\n",
    "\n",
    "# remove first n_lags Y values since they are not really predicted\n",
    "Y_hat = Y_hat[n_lags:,:]\n",
    "    \n",
    "Y_hat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Using statsmodels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "var_statsmodel = VAR(data)\n",
    "result = var_statsmodel.fit(n_lags)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.00106008],\n",
       "       [0.00015251],\n",
       "       [0.04907764]])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.coefs_exog # coefs for const"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[ 8.18738147e-02, -4.09705267e-03,  1.31690824e-04],\n",
       "        [ 2.64177843e-01,  7.60173471e-02,  6.26504703e-03],\n",
       "        [-3.13674354e+00,  2.77190854e+00,  1.31160993e+00]],\n",
       "\n",
       "       [[ 1.52147938e-01,  6.99007077e-04, -8.27967242e-06],\n",
       "        [-1.22091956e-01,  1.52766893e-01, -5.92688710e-03],\n",
       "        [-2.02144795e+01,  2.21810425e-01, -3.18219956e-01]]])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.coefs # coefs for lagged variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "  Summary of Regression Results   \n",
       "==================================\n",
       "Model:                         VAR\n",
       "Method:                        OLS\n",
       "Date:           Fri, 01, May, 2020\n",
       "Time:                     19:06:19\n",
       "--------------------------------------------------------------------\n",
       "No. of Equations:         3.00000    BIC:                   -27.2942\n",
       "Nobs:                     338.000    HQIC:                  -27.4370\n",
       "Log likelihood:           3235.05    FPE:                1.10446e-12\n",
       "AIC:                     -27.5317    Det(Omega_mle):     1.03859e-12\n",
       "--------------------------------------------------------------------\n",
       "Results for equation y1\n",
       "========================================================================\n",
       "           coefficient       std. error           t-stat            prob\n",
       "------------------------------------------------------------------------\n",
       "const         0.001060         0.000126            8.432           0.000\n",
       "L1.y1         0.081874         0.054707            1.497           0.134\n",
       "L1.y2        -0.004097         0.006925           -0.592           0.554\n",
       "L1.y3         0.000132         0.000212            0.622           0.534\n",
       "L2.y1         0.152148         0.053906            2.822           0.005\n",
       "L2.y2         0.000699         0.006872            0.102           0.919\n",
       "L2.y3        -0.000008         0.000210           -0.039           0.969\n",
       "========================================================================\n",
       "\n",
       "Results for equation y2\n",
       "========================================================================\n",
       "           coefficient       std. error           t-stat            prob\n",
       "------------------------------------------------------------------------\n",
       "const         0.000153         0.000979            0.156           0.876\n",
       "L1.y1         0.264178         0.425998            0.620           0.535\n",
       "L1.y2         0.076017         0.053924            1.410           0.159\n",
       "L1.y3         0.006265         0.001648            3.802           0.000\n",
       "L2.y1        -0.122092         0.419761           -0.291           0.771\n",
       "L2.y2         0.152767         0.053513            2.855           0.004\n",
       "L2.y3        -0.005927         0.001639           -3.617           0.000\n",
       "========================================================================\n",
       "\n",
       "Results for equation y3\n",
       "========================================================================\n",
       "           coefficient       std. error           t-stat            prob\n",
       "------------------------------------------------------------------------\n",
       "const         0.049078         0.031354            1.565           0.118\n",
       "L1.y1        -3.136744        13.643520           -0.230           0.818\n",
       "L1.y2         2.771909         1.727042            1.605           0.108\n",
       "L1.y3         1.311610         0.052778           24.852           0.000\n",
       "L2.y1       -20.214480        13.443747           -1.504           0.133\n",
       "L2.y2         0.221810         1.713878            0.129           0.897\n",
       "L2.y3        -0.318220         0.052484           -6.063           0.000\n",
       "========================================================================\n",
       "\n",
       "Correlation matrix of residuals\n",
       "            y1        y2        y3\n",
       "y1    1.000000 -0.041667  0.015543\n",
       "y2   -0.041667  1.000000  0.126311\n",
       "y3    0.015543  0.126311  1.000000\n",
       "\n"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/statsmodels/base/wrapper.py:36: FutureWarning: y is a deprecated alias for endog, will be removed in version 0.11.0\n",
      "  obj = getattr(results, attr)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[ 1.53906927e-03, -8.53453233e-04,  2.15852193e+00],\n",
       "       [ 1.45554085e-03,  9.19046652e-04,  2.16975697e+00],\n",
       "       [ 1.67691545e-03,  1.08892639e-03,  2.17474872e+00],\n",
       "       [ 1.68343888e-03,  1.40598473e-03,  2.17957889e+00],\n",
       "       [ 1.71706917e-03,  1.43140957e-03,  2.18074681e+00]])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.forecast(result.y, steps=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.53906927e-03,  5.41070604e-05,  2.11050437e+00],\n",
       "       [ 6.31508974e-04, -8.53453233e-04,  2.10959681e+00],\n",
       "       [ 4.95566303e-02,  4.80716681e-02,  2.15852193e+00]])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.coefs_exog + result.coefs[0] @ result.y[-1] + result.coefs[1] @ result.y[-2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### VAR with the 0 prior on coefficients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[ 1.28258999e-03,  3.64333906e-04,  2.38070216e-02],\n",
       "        [ 3.94943273e-06,  3.62101985e-06, -1.20819571e-04],\n",
       "        [-5.73616183e-06,  1.63438233e-04,  8.45511272e-03],\n",
       "        [ 8.96931594e-05,  3.37949564e-03,  8.59791455e-01],\n",
       "        [ 5.42428792e-06, -4.99541700e-06, -5.44668322e-04],\n",
       "        [-1.69595472e-06,  2.52597869e-04,  4.31367046e-03],\n",
       "        [ 9.19178361e-05, -2.90708065e-03,  1.26935082e-01]])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# OLS estimation\n",
    "penalty = 10\n",
    "bias_matrix = penalty*np.identity(X.shape[1])\n",
    "\n",
    "beta_array_biased = np.linalg.inv(X.T @ X + bias_matrix) @ X.T @ Y\n",
    "beta_array_biased"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[1.71759234e-03, 1.46377852e-03, 2.38332317e+00]])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Forecast 1 period ahead\n",
    "np.asmatrix(X[-1]) @ np.asmatrix(beta_array_biased)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating a class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "class VAR_Model():\n",
    "    \"\"\"\n",
    "    Handcrafted VAR model allowing for Ridge penalty.\n",
    "    \"\"\"\n",
    "    \n",
    "    def __init__(self, data):\n",
    "        self.data = data\n",
    "       \n",
    "    \n",
    "    def fit(self, n_lags, include_const=True, ridge_penalty=0):\n",
    "        \n",
    "        import numpy as np\n",
    "        \n",
    "        self.n_lags = n_lags\n",
    "        self.include_const = include_const\n",
    "        self.n_variables = data.shape[1] + self.include_const\n",
    "        self.n_obs = data.shape[0]\n",
    "        self.ridge_penalty = ridge_penalty\n",
    "        \n",
    "        # transform data to X and Y\n",
    "        endog_variables_lst = [data[lag-1:-n_lags+lag-1] for lag in range(1,n_lags+1)][::-1]\n",
    "        X = np.hstack(endog_variables_lst)\n",
    "        if include_const == True:\n",
    "            X = np.append(np.ones((self.n_obs-n_lags,1)), X, axis=1)\n",
    "        self.X = X\n",
    "        Y = data[n_lags:]\n",
    "        self.Y = Y\n",
    "        \n",
    "        # Ridge penalty\n",
    "        bias_matrix = ridge_penalty*np.identity(X.shape[1])\n",
    "        \n",
    "        # OLS estimation\n",
    "        beta_array = np.linalg.inv(X.T @ X + bias_matrix) @ X.T @ Y\n",
    "        self.beta_array = beta_array\n",
    "        \n",
    "    \n",
    "    def predict(self, Y, horizon):\n",
    "\n",
    "        # create new X for the first forecast horizon\n",
    "        X_new = np.hstack( [Y[-h] for h in range(1,n_lags+1)] ) # # all needed X vectors stacked in a matrix\n",
    "        if include_const:\n",
    "            X_new = np.append(1, X_new)\n",
    "\n",
    "        X_hat = np.asmatrix(X_new) # since it is a first iteration\n",
    "\n",
    "        # last forecast (here it is number one)\n",
    "        Y_hat_last = np.asmatrix(X_hat[-1]) @ np.asmatrix(beta_array)\n",
    "        Y_hat = np.asmatrix( np.vstack([Y[-n_lags:], Y_hat_last]) )\n",
    "\n",
    "        if horizon > 1:\n",
    "            for i in range(1, horizon):\n",
    "\n",
    "                # create new X now using Y_hat\n",
    "                X_new = np.hstack( [Y_hat[-h] for h in range(1,n_lags+1)] ) # # all needed X vectors stacked in a matrix\n",
    "                if include_const:\n",
    "                    X_new = np.append(1, X_new)\n",
    "\n",
    "                X_hat = np.vstack([X_hat, X_new]) # add current X to X_hat\n",
    "\n",
    "                # forecast using the last X_hat\n",
    "                Y_hat_last = np.asmatrix(X_hat[-1]) @ np.asmatrix(beta_array)\n",
    "                Y_hat = np.asmatrix( np.vstack([Y_hat, Y_hat_last]) )\n",
    "\n",
    "        # remove first n_lags Y values since they are not really predicted\n",
    "        Y_hat = Y_hat[n_lags:,:]\n",
    "                \n",
    "        return Y_hat\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "var_model = VAR_Model(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[ 1.06007528e-03,  1.52514982e-04,  4.90776363e-02],\n",
       "        [ 8.18738147e-02,  2.64177843e-01, -3.13674354e+00],\n",
       "        [-4.09705267e-03,  7.60173471e-02,  2.77190854e+00],\n",
       "        [ 1.31690824e-04,  6.26504703e-03,  1.31160993e+00],\n",
       "        [ 1.52147938e-01, -1.22091956e-01, -2.02144795e+01],\n",
       "        [ 6.99007077e-04,  1.52766893e-01,  2.21810425e-01],\n",
       "        [-8.27967242e-06, -5.92688710e-03, -3.18219956e-01]])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "var_model.fit(2)\n",
    "var_model.beta_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = var_model.predict(Y, 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[ 1.53906927e-03, -8.53453233e-04,  2.15852193e+00],\n",
       "        [ 1.45554085e-03,  9.19046652e-04,  2.16975697e+00],\n",
       "        [ 1.67691545e-03,  1.08892639e-03,  2.17474872e+00],\n",
       "        [ 1.68343888e-03,  1.40598473e-03,  2.17957889e+00],\n",
       "        [ 1.71706917e-03,  1.43140957e-03,  2.18074681e+00]])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.53906927e-03, -8.53453233e-04,  2.15852193e+00],\n",
       "       [ 1.45554085e-03,  9.19046652e-04,  2.16975697e+00],\n",
       "       [ 1.67691545e-03,  1.08892639e-03,  2.17474872e+00],\n",
       "       [ 1.68343888e-03,  1.40598473e-03,  2.17957889e+00],\n",
       "       [ 1.71706917e-03,  1.43140957e-03,  2.18074681e+00]])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.forecast(result.y, steps=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>CPI growth</th>\n",
       "      <th>Industrial production growth</th>\n",
       "      <th>Interest rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019M01</th>\n",
       "      <td>0.002396</td>\n",
       "      <td>-0.006711</td>\n",
       "      <td>2.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019M02</th>\n",
       "      <td>0.001104</td>\n",
       "      <td>-0.003378</td>\n",
       "      <td>2.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019M03</th>\n",
       "      <td>0.001475</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019M04</th>\n",
       "      <td>0.001381</td>\n",
       "      <td>-0.003390</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019M05</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         CPI growth  Industrial production growth  Interest rate\n",
       "2019M01    0.002396                     -0.006711           2.55\n",
       "2019M02    0.001104                     -0.003378           2.54\n",
       "2019M03    0.001475                      0.000000           2.40\n",
       "2019M04    0.001381                     -0.003390           2.39\n",
       "2019M05    0.000000                      0.000000           2.21"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x864 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, axes = plt.subplots(3, 1, sharex=True, figsize=(20,12))\n",
    "sns.lineplot(x=[*range(predictions.shape[0])], y=100*np.ravel(predictions[:,0]), ax=axes[0])\n",
    "axes[0].set_title('CPI growth, %')\n",
    "sns.lineplot(x=[*range(predictions.shape[0])], y=100*np.ravel(predictions[:,1]), ax=axes[1])\n",
    "axes[1].set_title('Industrial production growth, %')\n",
    "sns.lineplot(x=[*range(predictions.shape[0])], y=np.ravel(predictions[:,2]), ax=axes[2])\n",
    "axes[2].set_title('Interest rate, %')\n",
    "axes[2].set_xlabel('Months from now');"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}