{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Markov switching dynamic regression models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook provides an example of the use of Markov switching models in statsmodels to estimate dynamic regression models with changes in regime. It follows the examples in the Stata Markov switching documentation, which can be found at http://www.stata.com/manuals14/tsmswitch.pdf."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import statsmodels.api as sm\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import datetime\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Federal funds rate with switching intercept\n",
    "\n",
    "The first example models the federal funds rate as noise around a constant intercept, but where the intercept changes during different regimes. The model is simply:\n",
    "\n",
    "$$r_t = \\mu_{S_t} + \\varepsilon_t \\qquad \\varepsilon_t \\sim N(0, \\sigma^2)$$\n",
    "\n",
    "where $S_t \\in \\{0, 1\\}$, and the regime transitions according to\n",
    "\n",
    "$$ P(S_t = s_t | S_{t-1} = s_{t-1}) =\n",
    "\\begin{bmatrix}\n",
    "p_{00} & p_{10} \\\\\n",
    "1 - p_{00} & 1 - p_{10}\n",
    "\\end{bmatrix}\n",
    "$$\n",
    "\n",
    "We will estimate the parameters of this model by maximum likelihood: $p_{00}, p_{10}, \\mu_0, \\mu_1, \\sigma^2$.\n",
    "\n",
    "The data used in this example can be found at https://www.stata-press.com/data/r14/usmacro."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the federal funds rate data\n",
    "from statsmodels.tsa.regime_switching.tests.test_markov_regression import fedfunds\n",
    "dta_fedfunds = pd.Series(fedfunds, index=pd.date_range('1954-07-01', '2010-10-01', freq='QS'))\n",
    "\n",
    "# Plot the data\n",
    "dta_fedfunds.plot(title='Federal funds rate', figsize=(12,3))\n",
    "\n",
    "# Fit the model\n",
    "# (a switching mean is the default of the MarkovRegession model)\n",
    "mod_fedfunds = sm.tsa.MarkovRegression(dta_fedfunds, k_regimes=2)\n",
    "res_fedfunds = mod_fedfunds.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Markov Switching Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>    <td>226</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>MarkovRegression</td> <th>  Log Likelihood     </th> <td>-508.636</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Fri, 24 Apr 2020</td> <th>  AIC                </th> <td>1027.272</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>14:17:04</td>     <th>  BIC                </th> <td>1044.375</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>             <td>07-01-1954</td>    <th>  HQIC               </th> <td>1034.174</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                   <td>- 10-01-2010</td>   <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>approx</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 0 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>    3.7088</td> <td>    0.177</td> <td>   20.988</td> <td> 0.000</td> <td>    3.362</td> <td>    4.055</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 1 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>    9.5568</td> <td>    0.300</td> <td>   31.857</td> <td> 0.000</td> <td>    8.969</td> <td>   10.145</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Non-switching parameters</caption>\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    4.4418</td> <td>    0.425</td> <td>   10.447</td> <td> 0.000</td> <td>    3.608</td> <td>    5.275</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime transition parameters</caption>\n",
       "<tr>\n",
       "     <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[0->0]</th> <td>    0.9821</td> <td>    0.010</td> <td>   94.443</td> <td> 0.000</td> <td>    0.962</td> <td>    1.002</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[1->0]</th> <td>    0.0504</td> <td>    0.027</td> <td>    1.876</td> <td> 0.061</td> <td>   -0.002</td> <td>    0.103</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using numerical (complex-step) differentiation."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                        Markov Switching Model Results                        \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                  226\n",
       "Model:               MarkovRegression   Log Likelihood                -508.636\n",
       "Date:                Fri, 24 Apr 2020   AIC                           1027.272\n",
       "Time:                        14:17:04   BIC                           1044.375\n",
       "Sample:                    07-01-1954   HQIC                          1034.174\n",
       "                         - 10-01-2010                                         \n",
       "Covariance Type:               approx                                         \n",
       "                             Regime 0 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          3.7088      0.177     20.988      0.000       3.362       4.055\n",
       "                             Regime 1 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          9.5568      0.300     31.857      0.000       8.969      10.145\n",
       "                           Non-switching parameters                           \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "sigma2         4.4418      0.425     10.447      0.000       3.608       5.275\n",
       "                         Regime transition parameters                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "p[0->0]        0.9821      0.010     94.443      0.000       0.962       1.002\n",
       "p[1->0]        0.0504      0.027      1.876      0.061      -0.002       0.103\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using numerical (complex-step) differentiation.\n",
       "\"\"\""
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_fedfunds.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the summary output, the mean federal funds rate in the first regime (the \"low regime\") is estimated to be $3.7$ whereas in the \"high regime\" it is $9.6$. Below we plot the smoothed probabilities of being in the high regime. The model suggests that the 1980's was a time-period in which a high federal funds rate existed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "res_fedfunds.smoothed_marginal_probabilities[1].plot(\n",
    "    title='Probability of being in the high regime', figsize=(12,3));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the estimated transition matrix we can calculate the expected duration of a low regime versus a high regime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[55.85400626 19.85506546]\n"
     ]
    }
   ],
   "source": [
    "print(res_fedfunds.expected_durations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A low regime is expected to persist for about fourteen years, whereas the high regime is expected to persist for only about five years."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Federal funds rate with switching intercept and lagged dependent variable\n",
    "\n",
    "The second example augments the previous model to include the lagged value of the federal funds rate.\n",
    "\n",
    "$$r_t = \\mu_{S_t} + r_{t-1} \\beta_{S_t} + \\varepsilon_t \\qquad \\varepsilon_t \\sim N(0, \\sigma^2)$$\n",
    "\n",
    "where $S_t \\in \\{0, 1\\}$, and the regime transitions according to\n",
    "\n",
    "$$ P(S_t = s_t | S_{t-1} = s_{t-1}) =\n",
    "\\begin{bmatrix}\n",
    "p_{00} & p_{10} \\\\\n",
    "1 - p_{00} & 1 - p_{10}\n",
    "\\end{bmatrix}\n",
    "$$\n",
    "\n",
    "We will estimate the parameters of this model by maximum likelihood: $p_{00}, p_{10}, \\mu_0, \\mu_1, \\beta_0, \\beta_1, \\sigma^2$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Fit the model\n",
    "mod_fedfunds2 = sm.tsa.MarkovRegression(\n",
    "    dta_fedfunds.iloc[1:], k_regimes=2, exog=dta_fedfunds.iloc[:-1])\n",
    "res_fedfunds2 = mod_fedfunds2.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Markov Switching Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>    <td>225</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>MarkovRegression</td> <th>  Log Likelihood     </th> <td>-264.711</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Fri, 24 Apr 2020</td> <th>  AIC                </th>  <td>543.421</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>14:17:04</td>     <th>  BIC                </th>  <td>567.334</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>             <td>10-01-1954</td>    <th>  HQIC               </th>  <td>553.073</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                   <td>- 10-01-2010</td>   <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>approx</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 0 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>    0.7245</td> <td>    0.289</td> <td>    2.510</td> <td> 0.012</td> <td>    0.159</td> <td>    1.290</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    0.7631</td> <td>    0.034</td> <td>   22.629</td> <td> 0.000</td> <td>    0.697</td> <td>    0.829</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 1 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>   -0.0989</td> <td>    0.118</td> <td>   -0.835</td> <td> 0.404</td> <td>   -0.331</td> <td>    0.133</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    1.0612</td> <td>    0.019</td> <td>   57.351</td> <td> 0.000</td> <td>    1.025</td> <td>    1.097</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Non-switching parameters</caption>\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    0.4783</td> <td>    0.050</td> <td>    9.642</td> <td> 0.000</td> <td>    0.381</td> <td>    0.576</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime transition parameters</caption>\n",
       "<tr>\n",
       "     <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[0->0]</th> <td>    0.6378</td> <td>    0.120</td> <td>    5.304</td> <td> 0.000</td> <td>    0.402</td> <td>    0.874</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[1->0]</th> <td>    0.1306</td> <td>    0.050</td> <td>    2.634</td> <td> 0.008</td> <td>    0.033</td> <td>    0.228</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using numerical (complex-step) differentiation."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                        Markov Switching Model Results                        \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                  225\n",
       "Model:               MarkovRegression   Log Likelihood                -264.711\n",
       "Date:                Fri, 24 Apr 2020   AIC                            543.421\n",
       "Time:                        14:17:04   BIC                            567.334\n",
       "Sample:                    10-01-1954   HQIC                           553.073\n",
       "                         - 10-01-2010                                         \n",
       "Covariance Type:               approx                                         \n",
       "                             Regime 0 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          0.7245      0.289      2.510      0.012       0.159       1.290\n",
       "x1             0.7631      0.034     22.629      0.000       0.697       0.829\n",
       "                             Regime 1 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const         -0.0989      0.118     -0.835      0.404      -0.331       0.133\n",
       "x1             1.0612      0.019     57.351      0.000       1.025       1.097\n",
       "                           Non-switching parameters                           \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "sigma2         0.4783      0.050      9.642      0.000       0.381       0.576\n",
       "                         Regime transition parameters                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "p[0->0]        0.6378      0.120      5.304      0.000       0.402       0.874\n",
       "p[1->0]        0.1306      0.050      2.634      0.008       0.033       0.228\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using numerical (complex-step) differentiation.\n",
       "\"\"\""
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_fedfunds2.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several things to notice from the summary output:\n",
    "\n",
    "1. The information criteria have decreased substantially, indicating that this model has a better fit than the previous model.\n",
    "2. The interpretation of the regimes, in terms of the intercept, have switched. Now the first regime has the higher intercept and the second regime has a lower intercept.\n",
    "\n",
    "Examining the smoothed probabilities of the high regime state, we now see quite a bit more variability."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "res_fedfunds2.smoothed_marginal_probabilities[0].plot(\n",
    "    title='Probability of being in the high regime', figsize=(12,3));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, the expected durations of each regime have decreased quite a bit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.76105188 7.65529154]\n"
     ]
    }
   ],
   "source": [
    "print(res_fedfunds2.expected_durations)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Taylor rule with 2 or 3 regimes\n",
    "\n",
    "We now include two additional exogenous variables - a measure of the output gap and a measure of inflation - to estimate a switching Taylor-type rule with both 2 and 3 regimes to see which fits the data better.\n",
    "\n",
    "Because the models can be often difficult to estimate, for the 3-regime model we employ a search over starting parameters to improve results, specifying 20 random search repetitions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Get the additional data\n",
    "from statsmodels.tsa.regime_switching.tests.test_markov_regression import ogap, inf\n",
    "dta_ogap = pd.Series(ogap, index=pd.date_range('1954-07-01', '2010-10-01', freq='QS'))\n",
    "dta_inf = pd.Series(inf, index=pd.date_range('1954-07-01', '2010-10-01', freq='QS'))\n",
    "\n",
    "exog = pd.concat((dta_fedfunds.shift(), dta_ogap, dta_inf), axis=1).iloc[4:]\n",
    "\n",
    "# Fit the 2-regime model\n",
    "mod_fedfunds3 = sm.tsa.MarkovRegression(\n",
    "    dta_fedfunds.iloc[4:], k_regimes=2, exog=exog)\n",
    "res_fedfunds3 = mod_fedfunds3.fit()\n",
    "\n",
    "# Fit the 3-regime model\n",
    "np.random.seed(12345)\n",
    "mod_fedfunds4 = sm.tsa.MarkovRegression(\n",
    "    dta_fedfunds.iloc[4:], k_regimes=3, exog=exog)\n",
    "res_fedfunds4 = mod_fedfunds4.fit(search_reps=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Markov Switching Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>    <td>222</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>MarkovRegression</td> <th>  Log Likelihood     </th> <td>-229.256</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Fri, 24 Apr 2020</td> <th>  AIC                </th>  <td>480.512</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>14:17:04</td>     <th>  BIC                </th>  <td>517.942</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>             <td>07-01-1955</td>    <th>  HQIC               </th>  <td>495.624</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                   <td>- 10-01-2010</td>   <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>approx</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 0 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>    0.6555</td> <td>    0.137</td> <td>    4.771</td> <td> 0.000</td> <td>    0.386</td> <td>    0.925</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    0.8314</td> <td>    0.033</td> <td>   24.951</td> <td> 0.000</td> <td>    0.766</td> <td>    0.897</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>    <td>    0.1355</td> <td>    0.029</td> <td>    4.609</td> <td> 0.000</td> <td>    0.078</td> <td>    0.193</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>    <td>   -0.0274</td> <td>    0.041</td> <td>   -0.671</td> <td> 0.502</td> <td>   -0.107</td> <td>    0.053</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 1 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>   -0.0945</td> <td>    0.128</td> <td>   -0.739</td> <td> 0.460</td> <td>   -0.345</td> <td>    0.156</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    0.9293</td> <td>    0.027</td> <td>   34.309</td> <td> 0.000</td> <td>    0.876</td> <td>    0.982</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>    <td>    0.0343</td> <td>    0.024</td> <td>    1.429</td> <td> 0.153</td> <td>   -0.013</td> <td>    0.081</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>    <td>    0.2125</td> <td>    0.030</td> <td>    7.147</td> <td> 0.000</td> <td>    0.154</td> <td>    0.271</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Non-switching parameters</caption>\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    0.3323</td> <td>    0.035</td> <td>    9.526</td> <td> 0.000</td> <td>    0.264</td> <td>    0.401</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime transition parameters</caption>\n",
       "<tr>\n",
       "     <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[0->0]</th> <td>    0.7279</td> <td>    0.093</td> <td>    7.828</td> <td> 0.000</td> <td>    0.546</td> <td>    0.910</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[1->0]</th> <td>    0.2115</td> <td>    0.064</td> <td>    3.298</td> <td> 0.001</td> <td>    0.086</td> <td>    0.337</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using numerical (complex-step) differentiation."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                        Markov Switching Model Results                        \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                  222\n",
       "Model:               MarkovRegression   Log Likelihood                -229.256\n",
       "Date:                Fri, 24 Apr 2020   AIC                            480.512\n",
       "Time:                        14:17:04   BIC                            517.942\n",
       "Sample:                    07-01-1955   HQIC                           495.624\n",
       "                         - 10-01-2010                                         \n",
       "Covariance Type:               approx                                         \n",
       "                             Regime 0 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          0.6555      0.137      4.771      0.000       0.386       0.925\n",
       "x1             0.8314      0.033     24.951      0.000       0.766       0.897\n",
       "x2             0.1355      0.029      4.609      0.000       0.078       0.193\n",
       "x3            -0.0274      0.041     -0.671      0.502      -0.107       0.053\n",
       "                             Regime 1 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const         -0.0945      0.128     -0.739      0.460      -0.345       0.156\n",
       "x1             0.9293      0.027     34.309      0.000       0.876       0.982\n",
       "x2             0.0343      0.024      1.429      0.153      -0.013       0.081\n",
       "x3             0.2125      0.030      7.147      0.000       0.154       0.271\n",
       "                           Non-switching parameters                           \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "sigma2         0.3323      0.035      9.526      0.000       0.264       0.401\n",
       "                         Regime transition parameters                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "p[0->0]        0.7279      0.093      7.828      0.000       0.546       0.910\n",
       "p[1->0]        0.2115      0.064      3.298      0.001       0.086       0.337\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using numerical (complex-step) differentiation.\n",
       "\"\"\""
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_fedfunds3.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3/dist-packages/statsmodels/base/model.py:1354: RuntimeWarning: invalid value encountered in sqrt\n",
      "  bse_ = np.sqrt(np.diag(self.cov_params()))\n",
      "/usr/lib/python3/dist-packages/scipy/stats/_distn_infrastructure.py:903: RuntimeWarning: invalid value encountered in greater\n",
      "  return (a < x) & (x < b)\n",
      "/usr/lib/python3/dist-packages/scipy/stats/_distn_infrastructure.py:903: RuntimeWarning: invalid value encountered in less\n",
      "  return (a < x) & (x < b)\n",
      "/usr/lib/python3/dist-packages/scipy/stats/_distn_infrastructure.py:1912: RuntimeWarning: invalid value encountered in less_equal\n",
      "  cond2 = cond0 & (x <= _a)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Markov Switching Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>    <td>222</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>MarkovRegression</td> <th>  Log Likelihood     </th> <td>-180.806</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Fri, 24 Apr 2020</td> <th>  AIC                </th>  <td>399.611</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>14:17:04</td>     <th>  BIC                </th>  <td>464.262</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>             <td>07-01-1955</td>    <th>  HQIC               </th>  <td>425.713</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                   <td>- 10-01-2010</td>   <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>approx</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 0 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>   -1.0250</td> <td>    0.290</td> <td>   -3.531</td> <td> 0.000</td> <td>   -1.594</td> <td>   -0.456</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    0.3277</td> <td>    0.086</td> <td>    3.812</td> <td> 0.000</td> <td>    0.159</td> <td>    0.496</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>    <td>    0.2036</td> <td>    0.049</td> <td>    4.152</td> <td> 0.000</td> <td>    0.107</td> <td>    0.300</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>    <td>    1.1381</td> <td>    0.081</td> <td>   13.977</td> <td> 0.000</td> <td>    0.978</td> <td>    1.298</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 1 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>   -0.0259</td> <td>    0.087</td> <td>   -0.298</td> <td> 0.765</td> <td>   -0.196</td> <td>    0.144</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    0.9737</td> <td>    0.019</td> <td>   50.265</td> <td> 0.000</td> <td>    0.936</td> <td>    1.012</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>    <td>    0.0341</td> <td>    0.017</td> <td>    2.030</td> <td> 0.042</td> <td>    0.001</td> <td>    0.067</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>    <td>    0.1215</td> <td>    0.022</td> <td>    5.606</td> <td> 0.000</td> <td>    0.079</td> <td>    0.164</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 2 parameters</caption>\n",
       "<tr>\n",
       "    <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th> <td>    0.7346</td> <td>    0.130</td> <td>    5.632</td> <td> 0.000</td> <td>    0.479</td> <td>    0.990</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>    <td>    0.8436</td> <td>    0.024</td> <td>   35.198</td> <td> 0.000</td> <td>    0.797</td> <td>    0.891</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x2</th>    <td>    0.1633</td> <td>    0.025</td> <td>    6.515</td> <td> 0.000</td> <td>    0.114</td> <td>    0.212</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x3</th>    <td>   -0.0499</td> <td>    0.027</td> <td>   -1.835</td> <td> 0.067</td> <td>   -0.103</td> <td>    0.003</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Non-switching parameters</caption>\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    0.1660</td> <td>    0.018</td> <td>    9.240</td> <td> 0.000</td> <td>    0.131</td> <td>    0.201</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime transition parameters</caption>\n",
       "<tr>\n",
       "     <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[0->0]</th> <td>    0.7214</td> <td>    0.117</td> <td>    6.177</td> <td> 0.000</td> <td>    0.493</td> <td>    0.950</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[1->0]</th> <td> 4.001e-08</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[2->0]</th> <td>    0.0783</td> <td>    0.038</td> <td>    2.079</td> <td> 0.038</td> <td>    0.004</td> <td>    0.152</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[0->1]</th> <td>    0.1044</td> <td>    0.095</td> <td>    1.103</td> <td> 0.270</td> <td>   -0.081</td> <td>    0.290</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[1->1]</th> <td>    0.8259</td> <td>    0.054</td> <td>   15.208</td> <td> 0.000</td> <td>    0.719</td> <td>    0.932</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[2->1]</th> <td>    0.2288</td> <td>    0.073</td> <td>    3.150</td> <td> 0.002</td> <td>    0.086</td> <td>    0.371</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using numerical (complex-step) differentiation."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                        Markov Switching Model Results                        \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                  222\n",
       "Model:               MarkovRegression   Log Likelihood                -180.806\n",
       "Date:                Fri, 24 Apr 2020   AIC                            399.611\n",
       "Time:                        14:17:04   BIC                            464.262\n",
       "Sample:                    07-01-1955   HQIC                           425.713\n",
       "                         - 10-01-2010                                         \n",
       "Covariance Type:               approx                                         \n",
       "                             Regime 0 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const         -1.0250      0.290     -3.531      0.000      -1.594      -0.456\n",
       "x1             0.3277      0.086      3.812      0.000       0.159       0.496\n",
       "x2             0.2036      0.049      4.152      0.000       0.107       0.300\n",
       "x3             1.1381      0.081     13.977      0.000       0.978       1.298\n",
       "                             Regime 1 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const         -0.0259      0.087     -0.298      0.765      -0.196       0.144\n",
       "x1             0.9737      0.019     50.265      0.000       0.936       1.012\n",
       "x2             0.0341      0.017      2.030      0.042       0.001       0.067\n",
       "x3             0.1215      0.022      5.606      0.000       0.079       0.164\n",
       "                             Regime 2 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          0.7346      0.130      5.632      0.000       0.479       0.990\n",
       "x1             0.8436      0.024     35.198      0.000       0.797       0.891\n",
       "x2             0.1633      0.025      6.515      0.000       0.114       0.212\n",
       "x3            -0.0499      0.027     -1.835      0.067      -0.103       0.003\n",
       "                           Non-switching parameters                           \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "sigma2         0.1660      0.018      9.240      0.000       0.131       0.201\n",
       "                         Regime transition parameters                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "p[0->0]        0.7214      0.117      6.177      0.000       0.493       0.950\n",
       "p[1->0]     4.001e-08        nan        nan        nan         nan         nan\n",
       "p[2->0]        0.0783      0.038      2.079      0.038       0.004       0.152\n",
       "p[0->1]        0.1044      0.095      1.103      0.270      -0.081       0.290\n",
       "p[1->1]        0.8259      0.054     15.208      0.000       0.719       0.932\n",
       "p[2->1]        0.2288      0.073      3.150      0.002       0.086       0.371\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using numerical (complex-step) differentiation.\n",
       "\"\"\""
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_fedfunds4.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Due to lower information criteria, we might prefer the 3-state model, with an interpretation of low-, medium-, and high-interest rate regimes. The smoothed probabilities of each regime are plotted below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(3, figsize=(10,7))\n",
    "\n",
    "ax = axes[0]\n",
    "ax.plot(res_fedfunds4.smoothed_marginal_probabilities[0])\n",
    "ax.set(title='Smoothed probability of a low-interest rate regime')\n",
    "\n",
    "ax = axes[1]\n",
    "ax.plot(res_fedfunds4.smoothed_marginal_probabilities[1])\n",
    "ax.set(title='Smoothed probability of a medium-interest rate regime')\n",
    "\n",
    "ax = axes[2]\n",
    "ax.plot(res_fedfunds4.smoothed_marginal_probabilities[2])\n",
    "ax.set(title='Smoothed probability of a high-interest rate regime')\n",
    "\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Switching variances\n",
    "\n",
    "We can also accommodate switching variances. In particular, we consider the model\n",
    "\n",
    "$$\n",
    "y_t = \\mu_{S_t} + y_{t-1} \\beta_{S_t} + \\varepsilon_t \\quad \\varepsilon_t \\sim N(0, \\sigma_{S_t}^2)\n",
    "$$\n",
    "\n",
    "We use maximum likelihood to estimate the parameters of this model: $p_{00}, p_{10}, \\mu_0, \\mu_1, \\beta_0, \\beta_1, \\sigma_0^2, \\sigma_1^2$.\n",
    "\n",
    "The application is to absolute returns on stocks, where the data can be found at https://www.stata-press.com/data/r14/snp500."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the federal funds rate data\n",
    "from statsmodels.tsa.regime_switching.tests.test_markov_regression import areturns\n",
    "dta_areturns = pd.Series(areturns, index=pd.date_range('2004-05-04', '2014-5-03', freq='W'))\n",
    "\n",
    "# Plot the data\n",
    "dta_areturns.plot(title='Absolute returns, S&P500', figsize=(12,3))\n",
    "\n",
    "# Fit the model\n",
    "mod_areturns = sm.tsa.MarkovRegression(\n",
    "    dta_areturns.iloc[1:], k_regimes=2, exog=dta_areturns.iloc[:-1], switching_variance=True)\n",
    "res_areturns = mod_areturns.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Markov Switching Model Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>           <td>y</td>        <th>  No. Observations:  </th>    <td>520</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>MarkovRegression</td> <th>  Log Likelihood     </th> <td>-745.798</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>            <td>Fri, 24 Apr 2020</td> <th>  AIC                </th> <td>1507.595</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                <td>14:17:04</td>     <th>  BIC                </th> <td>1541.626</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>             <td>05-16-2004</td>    <th>  HQIC               </th> <td>1520.926</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                   <td>- 04-27-2014</td>   <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>approx</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 0 parameters</caption>\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>  <td>    0.7641</td> <td>    0.078</td> <td>    9.761</td> <td> 0.000</td> <td>    0.611</td> <td>    0.918</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>     <td>    0.0791</td> <td>    0.030</td> <td>    2.620</td> <td> 0.009</td> <td>    0.020</td> <td>    0.138</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    0.3476</td> <td>    0.061</td> <td>    5.694</td> <td> 0.000</td> <td>    0.228</td> <td>    0.467</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime 1 parameters</caption>\n",
       "<tr>\n",
       "     <td></td>       <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>const</th>  <td>    1.9728</td> <td>    0.278</td> <td>    7.086</td> <td> 0.000</td> <td>    1.427</td> <td>    2.518</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>x1</th>     <td>    0.5280</td> <td>    0.086</td> <td>    6.155</td> <td> 0.000</td> <td>    0.360</td> <td>    0.696</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th> <td>    2.5771</td> <td>    0.405</td> <td>    6.357</td> <td> 0.000</td> <td>    1.783</td> <td>    3.372</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Regime transition parameters</caption>\n",
       "<tr>\n",
       "     <td></td>        <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[0->0]</th> <td>    0.7531</td> <td>    0.063</td> <td>   11.871</td> <td> 0.000</td> <td>    0.629</td> <td>    0.877</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>p[1->0]</th> <td>    0.6825</td> <td>    0.066</td> <td>   10.301</td> <td> 0.000</td> <td>    0.553</td> <td>    0.812</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using numerical (complex-step) differentiation."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                        Markov Switching Model Results                        \n",
       "==============================================================================\n",
       "Dep. Variable:                      y   No. Observations:                  520\n",
       "Model:               MarkovRegression   Log Likelihood                -745.798\n",
       "Date:                Fri, 24 Apr 2020   AIC                           1507.595\n",
       "Time:                        14:17:04   BIC                           1541.626\n",
       "Sample:                    05-16-2004   HQIC                          1520.926\n",
       "                         - 04-27-2014                                         \n",
       "Covariance Type:               approx                                         \n",
       "                             Regime 0 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          0.7641      0.078      9.761      0.000       0.611       0.918\n",
       "x1             0.0791      0.030      2.620      0.009       0.020       0.138\n",
       "sigma2         0.3476      0.061      5.694      0.000       0.228       0.467\n",
       "                             Regime 1 parameters                              \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "const          1.9728      0.278      7.086      0.000       1.427       2.518\n",
       "x1             0.5280      0.086      6.155      0.000       0.360       0.696\n",
       "sigma2         2.5771      0.405      6.357      0.000       1.783       3.372\n",
       "                         Regime transition parameters                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "p[0->0]        0.7531      0.063     11.871      0.000       0.629       0.877\n",
       "p[1->0]        0.6825      0.066     10.301      0.000       0.553       0.812\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using numerical (complex-step) differentiation.\n",
       "\"\"\""
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_areturns.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The first regime is a low-variance regime and the second regime is a high-variance regime. Below we plot the probabilities of being in the low-variance regime. Between 2008 and 2012 there does not appear to be a clear indication of one regime guiding the economy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "res_areturns.smoothed_marginal_probabilities[0].plot(\n",
    "    title='Probability of being in a low-variance regime', figsize=(12,3));"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}
