{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fitting the Future with time series analysis\n",
    "> What lies ahead in this chapter is you predicting what lies ahead in your data. You'll learn how to use the elegant statsmodels package to fit ARMA, ARIMA and ARMAX models. Then you'll use your models to predict the uncertain future of stock prices! This is the Summary of lecture \"ARIMA Models in Python\", via datacamp.\n",
    "\n",
    "- toc: true \n",
    "- badges: true\n",
    "- comments: true\n",
    "- author: Chanseok Kang\n",
    "- categories: [Python, Datacamp, Time_Series_Analysis]\n",
    "- image: images/arima_forecast.png"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (10, 5)\n",
    "plt.style.use('fivethirtyeight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting time series models\n",
    "- Introduction to ARMAX models\n",
    "    - Exogenous ARMA\n",
    "    - Use external variables as well as time series\n",
    "    - ARMAX = ARMA + linear regression\n",
    "- ARMAX equation\n",
    "    - ARMA(1, 1) model:\n",
    "$$ y_t = a_1 y_{t-1} + m_1 \\epsilon_{t-1} + \\epsilon_t $$\n",
    "    - ARMAX(1, 1) model:\n",
    "$$ y_t = x_1 z_t + a_1 + y_{t-1} + m_1 \\epsilon_{t-1} + \\epsilon_t $$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fitting AR and MA models\n",
    "In this exercise you will fit an AR and an MA model to some data. The data here has been generated using the ```arma_generate_sample()``` function we used before.\n",
    "\n",
    "You know the real AR and MA parameters used to create this data so it is a really good way to gain some confidence with ARMA models and know you are doing it right. In the next exercise you'll move onto some real world data with confidence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>timeseries_1</th>\n",
       "      <th>timeseries_2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.183108</td>\n",
       "      <td>-0.183108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.245540</td>\n",
       "      <td>-0.117365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.258830</td>\n",
       "      <td>-0.218789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.279635</td>\n",
       "      <td>-0.169041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.384736</td>\n",
       "      <td>-0.282374</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   timeseries_1  timeseries_2\n",
       "0     -0.183108     -0.183108\n",
       "1     -0.245540     -0.117365\n",
       "2     -0.258830     -0.218789\n",
       "3     -0.279635     -0.169041\n",
       "4     -0.384736     -0.282374"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample = pd.read_csv('./dataset/sample.csv', index_col=0)\n",
    "sample.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### AR(2) model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                              ARMA Model Results                              \n",
      "==============================================================================\n",
      "Dep. Variable:           timeseries_1   No. Observations:                 1000\n",
      "Model:                     ARMA(2, 0)   Log Likelihood                 148.855\n",
      "Method:                       css-mle   S.D. of innovations              0.208\n",
      "Date:                Mon, 15 Jun 2020   AIC                           -289.709\n",
      "Time:                        18:46:18   BIC                           -270.078\n",
      "Sample:                             0   HQIC                          -282.248\n",
      "                                                                              \n",
      "======================================================================================\n",
      "                         coef    std err          z      P>|z|      [0.025      0.975]\n",
      "--------------------------------------------------------------------------------------\n",
      "const                 -0.0027      0.018     -0.151      0.880      -0.037       0.032\n",
      "ar.L1.timeseries_1     0.8980      0.030     29.510      0.000       0.838       0.958\n",
      "ar.L2.timeseries_1    -0.2704      0.030     -8.884      0.000      -0.330      -0.211\n",
      "                                    Roots                                    \n",
      "=============================================================================\n",
      "                  Real          Imaginary           Modulus         Frequency\n",
      "-----------------------------------------------------------------------------\n",
      "AR.1            1.6603           -0.9702j            1.9230           -0.0842\n",
      "AR.2            1.6603           +0.9702j            1.9230            0.0842\n",
      "-----------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "from statsmodels.tsa.arima_model import ARMA\n",
    "\n",
    "# Instantiate the model\n",
    "model = ARMA(sample['timeseries_1'], order=(2, 0))\n",
    "\n",
    "# Fit the model\n",
    "results = model.fit()\n",
    "\n",
    "# Print summary\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### MA(3) model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                              ARMA Model Results                              \n",
      "==============================================================================\n",
      "Dep. Variable:           timeseries_2   No. Observations:                 1000\n",
      "Model:                     ARMA(0, 3)   Log Likelihood                 149.007\n",
      "Method:                       css-mle   S.D. of innovations              0.208\n",
      "Date:                Mon, 15 Jun 2020   AIC                           -288.014\n",
      "Time:                        18:46:19   BIC                           -263.475\n",
      "Sample:                             0   HQIC                          -278.687\n",
      "                                                                              \n",
      "======================================================================================\n",
      "                         coef    std err          z      P>|z|      [0.025      0.975]\n",
      "--------------------------------------------------------------------------------------\n",
      "const                 -0.0018      0.012     -0.159      0.874      -0.024       0.021\n",
      "ma.L1.timeseries_2     0.1995      0.031      6.352      0.000       0.138       0.261\n",
      "ma.L2.timeseries_2     0.6359      0.028     22.718      0.000       0.581       0.691\n",
      "ma.L3.timeseries_2    -0.0833      0.029     -2.872      0.004      -0.140      -0.026\n",
      "                                    Roots                                    \n",
      "=============================================================================\n",
      "                  Real          Imaginary           Modulus         Frequency\n",
      "-----------------------------------------------------------------------------\n",
      "MA.1           -0.2389           -1.1928j            1.2165           -0.2815\n",
      "MA.2           -0.2389           +1.1928j            1.2165            0.2815\n",
      "MA.3            8.1089           -0.0000j            8.1089           -0.0000\n",
      "-----------------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# Instantiate the model\n",
    "model = ARMA(sample['timeseries_2'], order=(0, 3))\n",
    "\n",
    "# Fit the model\n",
    "results = model.fit()\n",
    "\n",
    "# Print summary\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fitting an ARMA model\n",
    "n this exercise you will fit an ARMA model to the earthquakes dataset. You saw before that the earthquakes dataset is stationary so you don't need to transform it at all. It comes ready for modeling straight out the ground. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>earthquakes_per_year</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1900-01-01</th>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1901-01-01</th>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1902-01-01</th>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1903-01-01</th>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1904-01-01</th>\n",
       "      <td>16.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            earthquakes_per_year\n",
       "date                            \n",
       "1900-01-01                  13.0\n",
       "1901-01-01                  14.0\n",
       "1902-01-01                   8.0\n",
       "1903-01-01                  10.0\n",
       "1904-01-01                  16.0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "earthquake = pd.read_csv('./dataset/earthquakes.csv', index_col='date', parse_dates=True)\n",
    "earthquake.drop(['Year'], axis=1, inplace=True)\n",
    "earthquake.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### ARMA(3, 1) model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                               ARMA Model Results                               \n",
      "================================================================================\n",
      "Dep. Variable:     earthquakes_per_year   No. Observations:                   99\n",
      "Model:                       ARMA(3, 1)   Log Likelihood                -315.673\n",
      "Method:                         css-mle   S.D. of innovations              5.853\n",
      "Date:                  Mon, 15 Jun 2020   AIC                            643.345\n",
      "Time:                          18:46:20   BIC                            658.916\n",
      "Sample:                      01-01-1900   HQIC                           649.645\n",
      "                           - 01-01-1998                                         \n",
      "==============================================================================================\n",
      "                                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "----------------------------------------------------------------------------------------------\n",
      "const                         19.6452      1.929     10.183      0.000      15.864      23.426\n",
      "ar.L1.earthquakes_per_year     0.5794      0.416      1.393      0.164      -0.236       1.394\n",
      "ar.L2.earthquakes_per_year     0.0251      0.208      0.121      0.904      -0.382       0.433\n",
      "ar.L3.earthquakes_per_year     0.1519      0.131      1.162      0.245      -0.104       0.408\n",
      "ma.L1.earthquakes_per_year    -0.1720      0.416     -0.413      0.679      -0.988       0.644\n",
      "                                    Roots                                    \n",
      "=============================================================================\n",
      "                  Real          Imaginary           Modulus         Frequency\n",
      "-----------------------------------------------------------------------------\n",
      "AR.1            1.2047           -0.0000j            1.2047           -0.0000\n",
      "AR.2           -0.6850           -2.2352j            2.3378           -0.2973\n",
      "AR.3           -0.6850           +2.2352j            2.3378            0.2973\n",
      "MA.1            5.8139           +0.0000j            5.8139            0.0000\n",
      "-----------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:162: ValueWarning: No frequency information was provided, so inferred frequency AS-JAN will be used.\n",
      "  % freq, ValueWarning)\n"
     ]
    }
   ],
   "source": [
    "# Instantiate the model\n",
    "model = ARMA(earthquake['earthquakes_per_year'], order=(3, 1))\n",
    "\n",
    "# Fit the model\n",
    "results = model.fit()\n",
    "\n",
    "# Print model fit summary\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fitting an ARMAX model\n",
    "In this exercise you will fit an ARMAX model to a time series which represents the wait times at an accident and emergency room for urgent medical care.\n",
    "\n",
    "The variable you would like to model is the wait times to be seen by a medical professional ```wait_times_hrs```. This may be related to an exogenous variable that you measured ```nurse_count``` which is the number of nurses on shift at any given time. These can be seen below.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>wait_times_hrs</th>\n",
       "      <th>nurse_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-03-04 00:00:00</th>\n",
       "      <td>1.747261</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-04 01:00:00</th>\n",
       "      <td>1.664634</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-04 02:00:00</th>\n",
       "      <td>1.647047</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-04 03:00:00</th>\n",
       "      <td>1.619512</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-04 04:00:00</th>\n",
       "      <td>1.480415</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     wait_times_hrs  nurse_count\n",
       "2019-03-04 00:00:00        1.747261          1.0\n",
       "2019-03-04 01:00:00        1.664634          1.0\n",
       "2019-03-04 02:00:00        1.647047          1.0\n",
       "2019-03-04 03:00:00        1.619512          1.0\n",
       "2019-03-04 04:00:00        1.480415          1.0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hospital = pd.read_csv('./dataset/hospital.csv', index_col=0, parse_dates=True)\n",
    "hospital.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "hospital.plot(subplots=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is a particularly interesting case of time series modeling as, if the number of nurses has an effect, you could change this to affect the wait times."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### ARMAX(2, 1) model to train on the ```wait_times_hrs``` using ```nurse_count``` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                              ARMA Model Results                              \n",
      "==============================================================================\n",
      "Dep. Variable:         wait_times_hrs   No. Observations:                  168\n",
      "Model:                     ARMA(2, 1)   Log Likelihood                 -11.834\n",
      "Method:                       css-mle   S.D. of innovations              0.259\n",
      "Date:                Mon, 15 Jun 2020   AIC                             35.668\n",
      "Time:                        18:46:22   BIC                             54.411\n",
      "Sample:                    03-04-2019   HQIC                            43.275\n",
      "                         - 03-10-2019                                         \n",
      "========================================================================================\n",
      "                           coef    std err          z      P>|z|      [0.025      0.975]\n",
      "----------------------------------------------------------------------------------------\n",
      "const                    2.1000      0.086     24.293      0.000       1.931       2.269\n",
      "nurse_count             -0.1171      0.013     -9.054      0.000      -0.142      -0.092\n",
      "ar.L1.wait_times_hrs     0.5693      0.164      3.468      0.001       0.248       0.891\n",
      "ar.L2.wait_times_hrs    -0.1612      0.131     -1.226      0.220      -0.419       0.096\n",
      "ma.L1.wait_times_hrs     0.3728      0.157      2.375      0.018       0.065       0.680\n",
      "                                    Roots                                    \n",
      "=============================================================================\n",
      "                  Real          Imaginary           Modulus         Frequency\n",
      "-----------------------------------------------------------------------------\n",
      "AR.1            1.7656           -1.7566j            2.4906           -0.1246\n",
      "AR.2            1.7656           +1.7566j            2.4906            0.1246\n",
      "MA.1           -2.6827           +0.0000j            2.6827            0.5000\n",
      "-----------------------------------------------------------------------------\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:162: ValueWarning: No frequency information was provided, so inferred frequency H will be used.\n",
      "  % freq, ValueWarning)\n"
     ]
    }
   ],
   "source": [
    "# Instantiate the model\n",
    "model = ARMA(hospital['wait_times_hrs'], order=(2, 1), exog=hospital['nurse_count'])\n",
    "\n",
    "# Fit the model\n",
    "results = model.fit()\n",
    "\n",
    "# Print model fit summary\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Forecasting\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generating one-step-ahead predictions\n",
    "It is very hard to forecast stock prices. Classic economics actually tells us that this should be impossible because of market clearing.\n",
    "\n",
    "Your task in this exercise is to attempt the impossible and predict the Amazon stock price anyway.\n",
    "\n",
    "In this exercise you will generate one-step-ahead predictions for the stock price as well as the uncertainty of these predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "amazon = pd.read_csv('./dataset/amazon_close.csv', parse_dates=True, index_col='date')\n",
    "amazon = amazon.iloc[::-1] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:218: ValueWarning: A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.\n",
      "  ' ignored when e.g. forecasting.', ValueWarning)\n",
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:218: ValueWarning: A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.\n",
      "  ' ignored when e.g. forecasting.', ValueWarning)\n",
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/base/model.py:568: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
      "  \"Check mle_retvals\", ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
    "\n",
    "model = SARIMAX(amazon.loc['2018-01-01':'2019-02-08'], order=(3, 1, 3), seasonal_order=(1, 0, 1, 7),\n",
    "                enforce_invertibility=False,\n",
    "                enforce_stationarity=False,\n",
    "                simple_differencing=False, \n",
    "                measurement_error=False,\n",
    "                k_trend=0)\n",
    "results = model.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>SARIMAX Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>                <td>close</td>              <th>  No. Observations:  </th>    <td>278</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>           <td>SARIMAX(3, 1, 3)x(1, 0, [1], 7)</td> <th>  Log Likelihood     </th> <td>-1338.384</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                   <td>Mon, 15 Jun 2020</td>         <th>  AIC                </th> <td>2694.769</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                       <td>18:46:24</td>             <th>  BIC                </th> <td>2727.020</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                         <td>0</td>                <th>  HQIC               </th> <td>2707.726</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                             <td> - 278</td>              <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>               <td>opg</td>               <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\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>ar.L1</th>   <td>    0.1074</td> <td>    0.048</td> <td>    2.258</td> <td> 0.024</td> <td>    0.014</td> <td>    0.201</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L2</th>   <td>    0.0521</td> <td>    0.038</td> <td>    1.359</td> <td> 0.174</td> <td>   -0.023</td> <td>    0.127</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.L3</th>   <td>   -0.8974</td> <td>    0.042</td> <td>  -21.606</td> <td> 0.000</td> <td>   -0.979</td> <td>   -0.816</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L1</th>   <td>   -0.1125</td> <td>    0.036</td> <td>   -3.113</td> <td> 0.002</td> <td>   -0.183</td> <td>   -0.042</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L2</th>   <td>   -0.1496</td> <td>    0.041</td> <td>   -3.671</td> <td> 0.000</td> <td>   -0.229</td> <td>   -0.070</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.L3</th>   <td>    0.9763</td> <td>    0.032</td> <td>   30.611</td> <td> 0.000</td> <td>    0.914</td> <td>    1.039</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ar.S.L7</th> <td>    0.1821</td> <td>    0.675</td> <td>    0.270</td> <td> 0.787</td> <td>   -1.141</td> <td>    1.506</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>ma.S.L7</th> <td>   -0.2247</td> <td>    0.666</td> <td>   -0.337</td> <td> 0.736</td> <td>   -1.531</td> <td>    1.081</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sigma2</th>  <td> 1319.0972</td> <td>   99.104</td> <td>   13.310</td> <td> 0.000</td> <td> 1124.858</td> <td> 1513.337</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Ljung-Box (Q):</th>          <td>28.84</td> <th>  Jarque-Bera (JB):  </th> <td>22.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Q):</th>                <td>0.91</td>  <th>  Prob(JB):          </th> <td>0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Heteroskedasticity (H):</th> <td>3.10</td>  <th>  Skew:              </th> <td>-0.35</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(H) (two-sided):</th>    <td>0.00</td>  <th>  Kurtosis:          </th> <td>4.23</td> \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix calculated using the outer product of gradients (complex-step)."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                      SARIMAX Results                                      \n",
       "===========================================================================================\n",
       "Dep. Variable:                               close   No. Observations:                  278\n",
       "Model:             SARIMAX(3, 1, 3)x(1, 0, [1], 7)   Log Likelihood               -1338.384\n",
       "Date:                             Mon, 15 Jun 2020   AIC                           2694.769\n",
       "Time:                                     18:46:24   BIC                           2727.020\n",
       "Sample:                                          0   HQIC                          2707.726\n",
       "                                             - 278                                         \n",
       "Covariance Type:                               opg                                         \n",
       "==============================================================================\n",
       "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "ar.L1          0.1074      0.048      2.258      0.024       0.014       0.201\n",
       "ar.L2          0.0521      0.038      1.359      0.174      -0.023       0.127\n",
       "ar.L3         -0.8974      0.042    -21.606      0.000      -0.979      -0.816\n",
       "ma.L1         -0.1125      0.036     -3.113      0.002      -0.183      -0.042\n",
       "ma.L2         -0.1496      0.041     -3.671      0.000      -0.229      -0.070\n",
       "ma.L3          0.9763      0.032     30.611      0.000       0.914       1.039\n",
       "ar.S.L7        0.1821      0.675      0.270      0.787      -1.141       1.506\n",
       "ma.S.L7       -0.2247      0.666     -0.337      0.736      -1.531       1.081\n",
       "sigma2      1319.0972     99.104     13.310      0.000    1124.858    1513.337\n",
       "===================================================================================\n",
       "Ljung-Box (Q):                       28.84   Jarque-Bera (JB):                22.02\n",
       "Prob(Q):                              0.91   Prob(JB):                         0.00\n",
       "Heteroskedasticity (H):               3.10   Skew:                            -0.35\n",
       "Prob(H) (two-sided):                  0.00   Kurtosis:                         4.23\n",
       "===================================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
       "\"\"\""
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1475.3973982  1462.84752096 1470.99540557 1498.12115484 1537.51838107\n",
      " 1508.09054892 1581.14926322 1627.24259216 1650.12797834 1649.53776158\n",
      " 1657.7163931  1648.12485235 1625.78085085 1671.04311494 1672.23342965\n",
      " 1683.43565237 1693.6949342  1642.5733451  1657.25345019 1652.28661236\n",
      " 1661.06421713 1620.90897063 1594.76080937 1679.5496602  1724.90278402\n",
      " 1629.30624018 1638.13065893 1647.51124676 1636.55265666 1606.68029738]\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions\n",
    "one_step_forecast = results.get_prediction(start=-30)\n",
    "\n",
    "# Extract prediction mean\n",
    "mean_forecast = one_step_forecast.predicted_mean\n",
    "\n",
    "# Get confidence intervals of predictions\n",
    "confidence_intervals = one_step_forecast.conf_int()\n",
    "\n",
    "# Select lower and upper confidence limits\n",
    "lower_limits = confidence_intervals.loc[:,'lower close']\n",
    "upper_limits = confidence_intervals.loc[:,'upper close']\n",
    "\n",
    "# Print best estimate  predictions\n",
    "print(mean_forecast.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting one-step-ahead predictions\n",
    "Now that you have your predictions on the Amazon stock, you should plot these predictions to see how you've done.\n",
    "\n",
    "You made predictions over the latest 30 days of data available, always forecasting just one day ahead. By evaluating these predictions you can judge how the model performs in making predictions for just the next day, where you don't know the answer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the amazon data\n",
    "plt.plot(amazon.index, amazon['close'], label='observed');\n",
    "\n",
    "# Plot your mean predictions\n",
    "plt.plot(mean_forecast.index, mean_forecast, color='r', label='forecast');\n",
    "\n",
    "# shade the area between your confidence limits\n",
    "plt.fill_between(lower_limits.index, lower_limits, upper_limits, color='pink');\n",
    "\n",
    "# Set labels, legends\n",
    "plt.xlabel('Date');\n",
    "plt.ylabel('Amazon Stock Price - Close USD');\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generating dynamic forecasts\n",
    "Now lets move a little further into the future, to dynamic predictions. What if you wanted to predict the Amazon stock price, not just for tomorrow, but for next week or next month? This is where dynamical predictions come in.\n",
    "\n",
    "Remember that in the video you learned how it is more difficult to make precise long-term forecasts because the shock terms add up. The further into the future the predictions go, the more uncertain. This is especially true with stock data and so you will likely find that your predictions in this exercise are not as precise as those in the last exercise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1475.3973982  1476.40017466 1468.3901068  1467.14976272 1468.18656451\n",
      " 1478.14922455 1476.87176087 1480.11773026 1472.62955383 1469.88023553\n",
      " 1466.93794607 1473.65129549 1477.18299701 1479.91259781 1475.05662359\n",
      " 1471.72120655 1468.06637732 1471.97702823 1475.28213547 1479.2113115\n",
      " 1476.17980423 1473.21901888 1469.25578864 1471.28791358 1473.97822619\n",
      " 1477.94469229 1476.70394567 1474.34205877 1470.48718766 1471.07043809]\n"
     ]
    }
   ],
   "source": [
    "# Generate predictions\n",
    "dynamic_forecast = results.get_prediction(start=-30, dynamic=True)\n",
    "\n",
    "# Extract prediction mean\n",
    "mean_forecast = dynamic_forecast.predicted_mean\n",
    "\n",
    "# Get confidence intervals of predictions\n",
    "confidence_intervals = dynamic_forecast.conf_int()\n",
    "\n",
    "# Select lower and upper confidence limits\n",
    "lower_limits = confidence_intervals.loc[:, 'lower close']\n",
    "upper_limits = confidence_intervals.loc[:, 'upper close']\n",
    "\n",
    "# Print bet estimate predictions\n",
    "print(mean_forecast.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting dynamic forecasts\n",
    "Time to plot your predictions. Remember that making dynamic predictions, means that your model makes predictions with no corrections, unlike the one-step-ahead predictions. This is kind of like making a forecast now for the next 30 days, and then waiting to see what happens before comparing how good your predictions were."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the amazon data\n",
    "plt.plot(amazon.index, amazon['close'], label='observed');\n",
    "\n",
    "# Plot your mean forecast\n",
    "plt.plot(mean_forecast.index, mean_forecast, label='forecast');\n",
    "\n",
    "# Shade the area between your confidence limits\n",
    "plt.fill_between(lower_limits.index, lower_limits, upper_limits, color='pink');\n",
    "\n",
    "# set labels, legends\n",
    "plt.xlabel('Date');\n",
    "plt.ylabel('Amazon Stock Price - Close USD');\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Intro to ARIMA models\n",
    "- The ARIMA model\n",
    "    - Take the difference\n",
    "    - Fit ARMA model\n",
    "    - Integrate forecast\n",
    "- ARIMA - Autoregressive Integrated Moving Average"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Differencing and fitting ARMA\n",
    "In this exercise you will fit an ARMA model to the Amazon stocks dataset. As you saw before, this is a non-stationary dataset. You will use differencing to make it stationary so that you can fit an ARMA model.\n",
    "\n",
    "In the next section you'll make a forecast of the differences and use this to forecast the actual values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:218: ValueWarning: A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.\n",
      "  ' ignored when e.g. forecasting.', ValueWarning)\n",
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:218: ValueWarning: A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.\n",
      "  ' ignored when e.g. forecasting.', ValueWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                               SARIMAX Results                                \n",
      "==============================================================================\n",
      "Dep. Variable:                  close   No. Observations:                 1258\n",
      "Model:               SARIMAX(2, 0, 2)   Log Likelihood               -5531.159\n",
      "Date:                Mon, 15 Jun 2020   AIC                          11072.319\n",
      "Time:                        18:46:30   BIC                          11098.005\n",
      "Sample:                             0   HQIC                         11081.972\n",
      "                               - 1258                                         \n",
      "Covariance Type:                  opg                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "ar.L1          1.0770      0.004    259.224      0.000       1.069       1.085\n",
      "ar.L2         -0.9950      0.004   -273.566      0.000      -1.002      -0.988\n",
      "ma.L1         -1.0915      0.006   -177.349      0.000      -1.104      -1.079\n",
      "ma.L2          0.9946      0.007    145.386      0.000       0.981       1.008\n",
      "sigma2       391.6344      6.804     57.556      0.000     378.298     404.971\n",
      "===================================================================================\n",
      "Ljung-Box (Q):                      104.43   Jarque-Bera (JB):              6865.60\n",
      "Prob(Q):                              0.00   Prob(JB):                         0.00\n",
      "Heteroskedasticity (H):              15.47   Skew:                            -0.20\n",
      "Prob(H) (two-sided):                  0.00   Kurtosis:                        14.44\n",
      "===================================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n"
     ]
    }
   ],
   "source": [
    "# Take the first difference of the data\n",
    "amazon_diff = amazon.diff().dropna()\n",
    "\n",
    "# Create ARMA(2, 2) model\n",
    "arma = SARIMAX(amazon_diff, order=(2, 0, 2))\n",
    "\n",
    "# Fit model\n",
    "arma_results = arma.fit()\n",
    "\n",
    "# Print fit summary\n",
    "print(arma_results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Unrolling ARMA forecast\n",
    "Now you will use the model that you trained in the previous exercise ```arma``` in order to forecast the absolute value of the Amazon stocks dataset. Remember that sometimes predicting the difference could be enough; will the stocks go up, or down; but sometimes the absolute value is key."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1258    1593.660837\n",
      "1259    1601.964525\n",
      "1260    1605.494152\n",
      "1261    1601.033697\n",
      "1262    1592.717948\n",
      "1263    1588.199887\n",
      "1264    1591.607802\n",
      "1265    1599.773421\n",
      "1266    1605.177031\n",
      "1267    1602.872224\n",
      "dtype: float64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:583: ValueWarning: No supported index is available. Prediction results will be given with an integer index beginning at `start`.\n",
      "  ValueWarning)\n"
     ]
    }
   ],
   "source": [
    "# Make arma forecast of next 10 differences\n",
    "arma_diff_forecast = arma_results.get_forecast(steps=10).predicted_mean\n",
    "\n",
    "# Integrate the difference forecast\n",
    "arma_int_forecast = np.cumsum(arma_diff_forecast)\n",
    "\n",
    "# Make absolute value forecast\n",
    "arma_value_forecast = arma_int_forecast + amazon.iloc[-1, 0]\n",
    "\n",
    "# Print forecast\n",
    "print(arma_value_forecast)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fitting an ARIMA model\n",
    "In this exercise you'll learn how to be lazy in time series modeling. Instead of taking the difference, modeling the difference and then integrating, you're just going to lets statsmodels do the hard work for you.\n",
    "\n",
    "You'll repeat the same exercise that you did before, of forecasting the absolute values of the Amazon stocks dataset, but this time with an ARIMA model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:218: ValueWarning: A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.\n",
      "  ' ignored when e.g. forecasting.', ValueWarning)\n",
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:218: ValueWarning: A date index has been provided, but it has no associated frequency information and so will be ignored when e.g. forecasting.\n",
      "  ' ignored when e.g. forecasting.', ValueWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1259    1593.662417\n",
      "1260    1601.932909\n",
      "1261    1605.426182\n",
      "1262    1600.960117\n",
      "1263    1592.674090\n",
      "1264    1588.192669\n",
      "1265    1591.609853\n",
      "1266    1599.749278\n",
      "1267    1605.116372\n",
      "1268    1602.799012\n",
      "dtype: float64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/chanseok/anaconda3/lib/python3.7/site-packages/statsmodels/tsa/base/tsa_model.py:583: ValueWarning: No supported index is available. Prediction results will be given with an integer index beginning at `start`.\n",
      "  ValueWarning)\n"
     ]
    }
   ],
   "source": [
    "# Create ARIMA(2, 1, 2) model\n",
    "arima = SARIMAX(amazon, order=(2, 1, 2))\n",
    "\n",
    "# Fit ARIMA model\n",
    "arima_results = arima.fit()\n",
    "\n",
    "# Make ARIMA forecast of next 10 values\n",
    "arima_value_forecast = arima_results.get_forecast(steps=10).predicted_mean\n",
    "\n",
    "# Print forecast\n",
    "print(arima_value_forecast)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the amazon data\n",
    "plt.plot(amazon.index[-100:], amazon.iloc[-100:]['close'], label='observed');\n",
    "\n",
    "# Plot your mean forecast\n",
    "rng = pd.date_range(start='2019-02-08', end='2019-02-21', freq='b')\n",
    "plt.plot(rng, arima_value_forecast.values, label='forecast');\n",
    "\n",
    "# Shade the area between your confidence limits\n",
    "# plt.fill_between(lower_limits.index, lower_limits, upper_limits, color='pink');\n",
    "\n",
    "# set labels, legends\n",
    "plt.xlabel('Date');\n",
    "plt.ylabel('Amazon Stock Price - Close USD');\n",
    "plt.legend();\n",
    "plt.savefig('../images/arima_forecast.png')"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}