11.SARIMA.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3/dist-packages/statsmodels/compat/pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.\n",
      "  from pandas.core import datetools\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import statsmodels.api as sm\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv('salesdata.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Date</th>\n",
       "      <th>Sales</th>\n",
       "      <th>Unnamed: 2</th>\n",
       "      <th>Unnamed: 3</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>01/01/07</td>\n",
       "      <td>4600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>=</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>02/01/07</td>\n",
       "      <td>5700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>03/01/07</td>\n",
       "      <td>6800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>04/01/07</td>\n",
       "      <td>7900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>05/01/07</td>\n",
       "      <td>9000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>06/01/07</td>\n",
       "      <td>10100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>07/01/07</td>\n",
       "      <td>5700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>08/01/07</td>\n",
       "      <td>6800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>09/01/07</td>\n",
       "      <td>7900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10/01/07</td>\n",
       "      <td>9000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11/01/07</td>\n",
       "      <td>10100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12/01/07</td>\n",
       "      <td>11200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>01/01/08</td>\n",
       "      <td>5800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>02/01/08</td>\n",
       "      <td>6900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>03/01/08</td>\n",
       "      <td>8000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>04/01/08</td>\n",
       "      <td>9100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>05/01/08</td>\n",
       "      <td>10200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>06/01/08</td>\n",
       "      <td>11300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>07/01/08</td>\n",
       "      <td>6900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>08/01/08</td>\n",
       "      <td>8000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>09/01/08</td>\n",
       "      <td>9100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>10/01/08</td>\n",
       "      <td>10200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>11/01/08</td>\n",
       "      <td>11300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>12/01/08</td>\n",
       "      <td>12400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>01/01/09</td>\n",
       "      <td>7000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>02/01/09</td>\n",
       "      <td>8100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>03/01/09</td>\n",
       "      <td>9200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>04/01/09</td>\n",
       "      <td>10300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>05/01/09</td>\n",
       "      <td>11400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>06/01/09</td>\n",
       "      <td>12500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>07/01/14</td>\n",
       "      <td>14100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>08/01/14</td>\n",
       "      <td>15200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>09/01/14</td>\n",
       "      <td>16300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>10/01/14</td>\n",
       "      <td>17400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>94</th>\n",
       "      <td>11/01/14</td>\n",
       "      <td>18500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>12/01/14</td>\n",
       "      <td>19600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>01/01/15</td>\n",
       "      <td>14200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>02/01/15</td>\n",
       "      <td>15300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>03/01/15</td>\n",
       "      <td>16400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>04/01/15</td>\n",
       "      <td>17500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>05/01/15</td>\n",
       "      <td>18600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101</th>\n",
       "      <td>06/01/15</td>\n",
       "      <td>19700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>07/01/15</td>\n",
       "      <td>15300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>08/01/15</td>\n",
       "      <td>16400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>104</th>\n",
       "      <td>09/01/15</td>\n",
       "      <td>17500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>10/01/15</td>\n",
       "      <td>18600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>11/01/15</td>\n",
       "      <td>19700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>12/01/15</td>\n",
       "      <td>20800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>01/01/16</td>\n",
       "      <td>15400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>02/01/16</td>\n",
       "      <td>16500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>03/01/16</td>\n",
       "      <td>17600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>04/01/16</td>\n",
       "      <td>18700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>05/01/16</td>\n",
       "      <td>19800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>06/01/16</td>\n",
       "      <td>20900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>07/01/16</td>\n",
       "      <td>16500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115</th>\n",
       "      <td>08/01/16</td>\n",
       "      <td>17600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>116</th>\n",
       "      <td>09/01/16</td>\n",
       "      <td>18700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>117</th>\n",
       "      <td>10/01/16</td>\n",
       "      <td>19800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>11/01/16</td>\n",
       "      <td>20900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>12/01/16</td>\n",
       "      <td>22000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>120 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date  Sales  Unnamed: 2  Unnamed: 3 Unnamed: 4\n",
       "0    01/01/07   4600         NaN         NaN          =\n",
       "1    02/01/07   5700         NaN         NaN        NaN\n",
       "2    03/01/07   6800         NaN         NaN        NaN\n",
       "3    04/01/07   7900         NaN         NaN        NaN\n",
       "4    05/01/07   9000         NaN         NaN        NaN\n",
       "5    06/01/07  10100         NaN         NaN        NaN\n",
       "6    07/01/07   5700         NaN         NaN        NaN\n",
       "7    08/01/07   6800         NaN         NaN        NaN\n",
       "8    09/01/07   7900         NaN         NaN        NaN\n",
       "9    10/01/07   9000         NaN         NaN        NaN\n",
       "10   11/01/07  10100         NaN         NaN        NaN\n",
       "11   12/01/07  11200         NaN         NaN        NaN\n",
       "12   01/01/08   5800         NaN         NaN        NaN\n",
       "13   02/01/08   6900         NaN         NaN        NaN\n",
       "14   03/01/08   8000         NaN         NaN        NaN\n",
       "15   04/01/08   9100         NaN         NaN        NaN\n",
       "16   05/01/08  10200         NaN         NaN        NaN\n",
       "17   06/01/08  11300         NaN         NaN        NaN\n",
       "18   07/01/08   6900         NaN         NaN        NaN\n",
       "19   08/01/08   8000         NaN         NaN        NaN\n",
       "20   09/01/08   9100         NaN         NaN        NaN\n",
       "21   10/01/08  10200         NaN         NaN        NaN\n",
       "22   11/01/08  11300         NaN         NaN        NaN\n",
       "23   12/01/08  12400         NaN         NaN        NaN\n",
       "24   01/01/09   7000         NaN         NaN        NaN\n",
       "25   02/01/09   8100         NaN         NaN        NaN\n",
       "26   03/01/09   9200         NaN         NaN        NaN\n",
       "27   04/01/09  10300         NaN         NaN        NaN\n",
       "28   05/01/09  11400         NaN         NaN        NaN\n",
       "29   06/01/09  12500         NaN         NaN        NaN\n",
       "..        ...    ...         ...         ...        ...\n",
       "90   07/01/14  14100         NaN         NaN        NaN\n",
       "91   08/01/14  15200         NaN         NaN        NaN\n",
       "92   09/01/14  16300         NaN         NaN        NaN\n",
       "93   10/01/14  17400         NaN         NaN        NaN\n",
       "94   11/01/14  18500         NaN         NaN        NaN\n",
       "95   12/01/14  19600         NaN         NaN        NaN\n",
       "96   01/01/15  14200         NaN         NaN        NaN\n",
       "97   02/01/15  15300         NaN         NaN        NaN\n",
       "98   03/01/15  16400         NaN         NaN        NaN\n",
       "99   04/01/15  17500         NaN         NaN        NaN\n",
       "100  05/01/15  18600         NaN         NaN        NaN\n",
       "101  06/01/15  19700         NaN         NaN        NaN\n",
       "102  07/01/15  15300         NaN         NaN        NaN\n",
       "103  08/01/15  16400         NaN         NaN        NaN\n",
       "104  09/01/15  17500         NaN         NaN        NaN\n",
       "105  10/01/15  18600         NaN         NaN        NaN\n",
       "106  11/01/15  19700         NaN         NaN        NaN\n",
       "107  12/01/15  20800         NaN         NaN        NaN\n",
       "108  01/01/16  15400         NaN         NaN        NaN\n",
       "109  02/01/16  16500         NaN         NaN        NaN\n",
       "110  03/01/16  17600         NaN         NaN        NaN\n",
       "111  04/01/16  18700         NaN         NaN        NaN\n",
       "112  05/01/16  19800         NaN         NaN        NaN\n",
       "113  06/01/16  20900         NaN         NaN        NaN\n",
       "114  07/01/16  16500         NaN         NaN        NaN\n",
       "115  08/01/16  17600         NaN         NaN        NaN\n",
       "116  09/01/16  18700         NaN         NaN        NaN\n",
       "117  10/01/16  19800         NaN         NaN        NaN\n",
       "118  11/01/16  20900         NaN         NaN        NaN\n",
       "119  12/01/16  22000         NaN         NaN        NaN\n",
       "\n",
       "[120 rows x 5 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.index=pd.to_datetime(df['Date'])"
   ]
  },
  {
   "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>Date</th>\n",
       "      <th>Sales</th>\n",
       "      <th>Unnamed: 2</th>\n",
       "      <th>Unnamed: 3</th>\n",
       "      <th>Unnamed: 4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2007-01-01</th>\n",
       "      <td>01/01/07</td>\n",
       "      <td>4600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>=</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-02-01</th>\n",
       "      <td>02/01/07</td>\n",
       "      <td>5700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-03-01</th>\n",
       "      <td>03/01/07</td>\n",
       "      <td>6800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-04-01</th>\n",
       "      <td>04/01/07</td>\n",
       "      <td>7900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-05-01</th>\n",
       "      <td>05/01/07</td>\n",
       "      <td>9000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-06-01</th>\n",
       "      <td>06/01/07</td>\n",
       "      <td>10100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-07-01</th>\n",
       "      <td>07/01/07</td>\n",
       "      <td>5700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-08-01</th>\n",
       "      <td>08/01/07</td>\n",
       "      <td>6800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-09-01</th>\n",
       "      <td>09/01/07</td>\n",
       "      <td>7900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-10-01</th>\n",
       "      <td>10/01/07</td>\n",
       "      <td>9000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-11-01</th>\n",
       "      <td>11/01/07</td>\n",
       "      <td>10100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007-12-01</th>\n",
       "      <td>12/01/07</td>\n",
       "      <td>11200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-01-01</th>\n",
       "      <td>01/01/08</td>\n",
       "      <td>5800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-02-01</th>\n",
       "      <td>02/01/08</td>\n",
       "      <td>6900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-03-01</th>\n",
       "      <td>03/01/08</td>\n",
       "      <td>8000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-04-01</th>\n",
       "      <td>04/01/08</td>\n",
       "      <td>9100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-05-01</th>\n",
       "      <td>05/01/08</td>\n",
       "      <td>10200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-06-01</th>\n",
       "      <td>06/01/08</td>\n",
       "      <td>11300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-07-01</th>\n",
       "      <td>07/01/08</td>\n",
       "      <td>6900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-08-01</th>\n",
       "      <td>08/01/08</td>\n",
       "      <td>8000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-09-01</th>\n",
       "      <td>09/01/08</td>\n",
       "      <td>9100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-10-01</th>\n",
       "      <td>10/01/08</td>\n",
       "      <td>10200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-11-01</th>\n",
       "      <td>11/01/08</td>\n",
       "      <td>11300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-12-01</th>\n",
       "      <td>12/01/08</td>\n",
       "      <td>12400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009-01-01</th>\n",
       "      <td>01/01/09</td>\n",
       "      <td>7000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009-02-01</th>\n",
       "      <td>02/01/09</td>\n",
       "      <td>8100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009-03-01</th>\n",
       "      <td>03/01/09</td>\n",
       "      <td>9200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009-04-01</th>\n",
       "      <td>04/01/09</td>\n",
       "      <td>10300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009-05-01</th>\n",
       "      <td>05/01/09</td>\n",
       "      <td>11400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009-06-01</th>\n",
       "      <td>06/01/09</td>\n",
       "      <td>12500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-07-01</th>\n",
       "      <td>07/01/14</td>\n",
       "      <td>14100</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-08-01</th>\n",
       "      <td>08/01/14</td>\n",
       "      <td>15200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-09-01</th>\n",
       "      <td>09/01/14</td>\n",
       "      <td>16300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-10-01</th>\n",
       "      <td>10/01/14</td>\n",
       "      <td>17400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-11-01</th>\n",
       "      <td>11/01/14</td>\n",
       "      <td>18500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014-12-01</th>\n",
       "      <td>12/01/14</td>\n",
       "      <td>19600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-01</th>\n",
       "      <td>01/01/15</td>\n",
       "      <td>14200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-02-01</th>\n",
       "      <td>02/01/15</td>\n",
       "      <td>15300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-03-01</th>\n",
       "      <td>03/01/15</td>\n",
       "      <td>16400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-04-01</th>\n",
       "      <td>04/01/15</td>\n",
       "      <td>17500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-05-01</th>\n",
       "      <td>05/01/15</td>\n",
       "      <td>18600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-06-01</th>\n",
       "      <td>06/01/15</td>\n",
       "      <td>19700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-07-01</th>\n",
       "      <td>07/01/15</td>\n",
       "      <td>15300</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-08-01</th>\n",
       "      <td>08/01/15</td>\n",
       "      <td>16400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-09-01</th>\n",
       "      <td>09/01/15</td>\n",
       "      <td>17500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-10-01</th>\n",
       "      <td>10/01/15</td>\n",
       "      <td>18600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-11-01</th>\n",
       "      <td>11/01/15</td>\n",
       "      <td>19700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-12-01</th>\n",
       "      <td>12/01/15</td>\n",
       "      <td>20800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-01-01</th>\n",
       "      <td>01/01/16</td>\n",
       "      <td>15400</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-02-01</th>\n",
       "      <td>02/01/16</td>\n",
       "      <td>16500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-01</th>\n",
       "      <td>03/01/16</td>\n",
       "      <td>17600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-04-01</th>\n",
       "      <td>04/01/16</td>\n",
       "      <td>18700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-05-01</th>\n",
       "      <td>05/01/16</td>\n",
       "      <td>19800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-06-01</th>\n",
       "      <td>06/01/16</td>\n",
       "      <td>20900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-07-01</th>\n",
       "      <td>07/01/16</td>\n",
       "      <td>16500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-08-01</th>\n",
       "      <td>08/01/16</td>\n",
       "      <td>17600</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-09-01</th>\n",
       "      <td>09/01/16</td>\n",
       "      <td>18700</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-10-01</th>\n",
       "      <td>10/01/16</td>\n",
       "      <td>19800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-01</th>\n",
       "      <td>11/01/16</td>\n",
       "      <td>20900</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01</th>\n",
       "      <td>12/01/16</td>\n",
       "      <td>22000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>120 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                Date  Sales  Unnamed: 2  Unnamed: 3 Unnamed: 4\n",
       "Date                                                          \n",
       "2007-01-01  01/01/07   4600         NaN         NaN          =\n",
       "2007-02-01  02/01/07   5700         NaN         NaN        NaN\n",
       "2007-03-01  03/01/07   6800         NaN         NaN        NaN\n",
       "2007-04-01  04/01/07   7900         NaN         NaN        NaN\n",
       "2007-05-01  05/01/07   9000         NaN         NaN        NaN\n",
       "2007-06-01  06/01/07  10100         NaN         NaN        NaN\n",
       "2007-07-01  07/01/07   5700         NaN         NaN        NaN\n",
       "2007-08-01  08/01/07   6800         NaN         NaN        NaN\n",
       "2007-09-01  09/01/07   7900         NaN         NaN        NaN\n",
       "2007-10-01  10/01/07   9000         NaN         NaN        NaN\n",
       "2007-11-01  11/01/07  10100         NaN         NaN        NaN\n",
       "2007-12-01  12/01/07  11200         NaN         NaN        NaN\n",
       "2008-01-01  01/01/08   5800         NaN         NaN        NaN\n",
       "2008-02-01  02/01/08   6900         NaN         NaN        NaN\n",
       "2008-03-01  03/01/08   8000         NaN         NaN        NaN\n",
       "2008-04-01  04/01/08   9100         NaN         NaN        NaN\n",
       "2008-05-01  05/01/08  10200         NaN         NaN        NaN\n",
       "2008-06-01  06/01/08  11300         NaN         NaN        NaN\n",
       "2008-07-01  07/01/08   6900         NaN         NaN        NaN\n",
       "2008-08-01  08/01/08   8000         NaN         NaN        NaN\n",
       "2008-09-01  09/01/08   9100         NaN         NaN        NaN\n",
       "2008-10-01  10/01/08  10200         NaN         NaN        NaN\n",
       "2008-11-01  11/01/08  11300         NaN         NaN        NaN\n",
       "2008-12-01  12/01/08  12400         NaN         NaN        NaN\n",
       "2009-01-01  01/01/09   7000         NaN         NaN        NaN\n",
       "2009-02-01  02/01/09   8100         NaN         NaN        NaN\n",
       "2009-03-01  03/01/09   9200         NaN         NaN        NaN\n",
       "2009-04-01  04/01/09  10300         NaN         NaN        NaN\n",
       "2009-05-01  05/01/09  11400         NaN         NaN        NaN\n",
       "2009-06-01  06/01/09  12500         NaN         NaN        NaN\n",
       "...              ...    ...         ...         ...        ...\n",
       "2014-07-01  07/01/14  14100         NaN         NaN        NaN\n",
       "2014-08-01  08/01/14  15200         NaN         NaN        NaN\n",
       "2014-09-01  09/01/14  16300         NaN         NaN        NaN\n",
       "2014-10-01  10/01/14  17400         NaN         NaN        NaN\n",
       "2014-11-01  11/01/14  18500         NaN         NaN        NaN\n",
       "2014-12-01  12/01/14  19600         NaN         NaN        NaN\n",
       "2015-01-01  01/01/15  14200         NaN         NaN        NaN\n",
       "2015-02-01  02/01/15  15300         NaN         NaN        NaN\n",
       "2015-03-01  03/01/15  16400         NaN         NaN        NaN\n",
       "2015-04-01  04/01/15  17500         NaN         NaN        NaN\n",
       "2015-05-01  05/01/15  18600         NaN         NaN        NaN\n",
       "2015-06-01  06/01/15  19700         NaN         NaN        NaN\n",
       "2015-07-01  07/01/15  15300         NaN         NaN        NaN\n",
       "2015-08-01  08/01/15  16400         NaN         NaN        NaN\n",
       "2015-09-01  09/01/15  17500         NaN         NaN        NaN\n",
       "2015-10-01  10/01/15  18600         NaN         NaN        NaN\n",
       "2015-11-01  11/01/15  19700         NaN         NaN        NaN\n",
       "2015-12-01  12/01/15  20800         NaN         NaN        NaN\n",
       "2016-01-01  01/01/16  15400         NaN         NaN        NaN\n",
       "2016-02-01  02/01/16  16500         NaN         NaN        NaN\n",
       "2016-03-01  03/01/16  17600         NaN         NaN        NaN\n",
       "2016-04-01  04/01/16  18700         NaN         NaN        NaN\n",
       "2016-05-01  05/01/16  19800         NaN         NaN        NaN\n",
       "2016-06-01  06/01/16  20900         NaN         NaN        NaN\n",
       "2016-07-01  07/01/16  16500         NaN         NaN        NaN\n",
       "2016-08-01  08/01/16  17600         NaN         NaN        NaN\n",
       "2016-09-01  09/01/16  18700         NaN         NaN        NaN\n",
       "2016-10-01  10/01/16  19800         NaN         NaN        NaN\n",
       "2016-11-01  11/01/16  20900         NaN         NaN        NaN\n",
       "2016-12-01  12/01/16  22000         NaN         NaN        NaN\n",
       "\n",
       "[120 rows x 5 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27d2ee9b00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Sales'].plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27d0d2fd68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(12,8))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27d0d2fd68>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "ax1 = fig.add_subplot(211)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f27d0da94a8>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ax1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = sm.graphics.tsa.plot_acf(df['Sales'].diff().dropna(), lags=40, ax=ax1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27d0d2fd68>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "ax2 = fig.add_subplot(212)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = sm.graphics.tsa.plot_pacf(df['Sales'].diff().dropna(), lags=40, ax=ax2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                 Statespace Model Results                                \n",
      "=========================================================================================\n",
      "Dep. Variable:                             Sales   No. Observations:                  120\n",
      "Model:             SARIMAX(0, 1, 0)x(0, 0, 1, 6)   Log Likelihood               -3157.530\n",
      "Date:                           Wed, 21 Aug 2019   AIC                           6321.060\n",
      "Time:                                   23:32:40   BIC                           6329.422\n",
      "Sample:                               01-01-2007   HQIC                          6324.456\n",
      "                                    - 12-01-2016                                         \n",
      "Covariance Type:                             opg                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "intercept    146.2185        nan        nan        nan         nan         nan\n",
      "ma.S.L6    -6.689e+14          0       -inf      0.000   -6.69e+14   -6.69e+14\n",
      "sigma2      3.964e-08   7.27e-09      5.452      0.000    2.54e-08    5.39e-08\n",
      "===================================================================================\n",
      "Ljung-Box (Q):                      715.62   Jarque-Bera (JB):                84.34\n",
      "Prob(Q):                              0.00   Prob(JB):                         0.00\n",
      "Heteroskedasticity (H):               1.00   Skew:                            -1.89\n",
      "Prob(H) (two-sided):                  1.00   Kurtosis:                         4.64\n",
      "===================================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n",
      "[2] Covariance matrix is singular or near-singular, with condition number    inf. Standard errors may be unstable.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python3/dist-packages/statsmodels/base/model.py:496: ConvergenceWarning: Maximum Likelihood optimization failed to converge. Check mle_retvals\n",
      "  \"Check mle_retvals\", ConvergenceWarning)\n",
      "/usr/lib/python3/dist-packages/statsmodels/base/model.py:1029: RuntimeWarning: invalid value encountered in sqrt\n",
      "  return np.sqrt(np.diag(self.cov_params()))\n",
      "/usr/lib/python3/dist-packages/statsmodels/base/model.py:1036: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  return self.params / self.bse\n",
      "/usr/lib/python3/dist-packages/statsmodels/tsa/statespace/mlemodel.py:1964: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  return self.params / self.bse\n",
      "/usr/lib/python3/dist-packages/scipy/stats/_distn_infrastructure.py:879: RuntimeWarning: invalid value encountered in greater\n",
      "  return (self.a < x) & (x < self.b)\n",
      "/usr/lib/python3/dist-packages/scipy/stats/_distn_infrastructure.py:879: RuntimeWarning: invalid value encountered in less\n",
      "  return (self.a < x) & (x < self.b)\n",
      "/usr/lib/python3/dist-packages/scipy/stats/_distn_infrastructure.py:1818: RuntimeWarning: invalid value encountered in less_equal\n",
      "  cond2 = cond0 & (x <= self.a)\n"
     ]
    }
   ],
   "source": [
    "model=sm.tsa.statespace.SARIMAX(endog=df['Sales'],order=(0,1,0),seasonal_order=(0,0,1,6),trend='c',enforce_invertibility=False)\n",
    "results=model.fit()\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2007-01-01    NaN\n",
      "2007-02-01    NaN\n",
      "2007-03-01    NaN\n",
      "2007-04-01    NaN\n",
      "2007-05-01    NaN\n",
      "2007-06-01    NaN\n",
      "2007-07-01    NaN\n",
      "2007-08-01    NaN\n",
      "2007-09-01    NaN\n",
      "2007-10-01    NaN\n",
      "2007-11-01    NaN\n",
      "2007-12-01    NaN\n",
      "2008-01-01    NaN\n",
      "2008-02-01    0.0\n",
      "2008-03-01    0.0\n",
      "2008-04-01    0.0\n",
      "2008-05-01    0.0\n",
      "2008-06-01    0.0\n",
      "2008-07-01    0.0\n",
      "2008-08-01    0.0\n",
      "2008-09-01    0.0\n",
      "2008-10-01    0.0\n",
      "2008-11-01    0.0\n",
      "2008-12-01    0.0\n",
      "2009-01-01    0.0\n",
      "2009-02-01    0.0\n",
      "2009-03-01    0.0\n",
      "2009-04-01    0.0\n",
      "2009-05-01    0.0\n",
      "2009-06-01    0.0\n",
      "             ... \n",
      "2014-07-01    0.0\n",
      "2014-08-01    0.0\n",
      "2014-09-01    0.0\n",
      "2014-10-01    0.0\n",
      "2014-11-01    0.0\n",
      "2014-12-01    0.0\n",
      "2015-01-01    0.0\n",
      "2015-02-01    0.0\n",
      "2015-03-01    0.0\n",
      "2015-04-01    0.0\n",
      "2015-05-01    0.0\n",
      "2015-06-01    0.0\n",
      "2015-07-01    0.0\n",
      "2015-08-01    0.0\n",
      "2015-09-01    0.0\n",
      "2015-10-01    0.0\n",
      "2015-11-01    0.0\n",
      "2015-12-01    0.0\n",
      "2016-01-01    0.0\n",
      "2016-02-01    0.0\n",
      "2016-03-01    0.0\n",
      "2016-04-01    0.0\n",
      "2016-05-01    0.0\n",
      "2016-06-01    0.0\n",
      "2016-07-01    0.0\n",
      "2016-08-01    0.0\n",
      "2016-09-01    0.0\n",
      "2016-10-01    0.0\n",
      "2016-11-01    0.0\n",
      "2016-12-01    0.0\n",
      "Name: Sales, Length: 120, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(df['Sales'].diff().diff(12))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Date\n",
      "2007-01-01       NaN\n",
      "2007-02-01       NaN\n",
      "2007-03-01       NaN\n",
      "2007-04-01       NaN\n",
      "2007-05-01       NaN\n",
      "2007-06-01       NaN\n",
      "2007-07-01       NaN\n",
      "2007-08-01       NaN\n",
      "2007-09-01       NaN\n",
      "2007-10-01       NaN\n",
      "2007-11-01       NaN\n",
      "2007-12-01       NaN\n",
      "2008-01-01    1200.0\n",
      "2008-02-01    1200.0\n",
      "2008-03-01    1200.0\n",
      "2008-04-01    1200.0\n",
      "2008-05-01    1200.0\n",
      "2008-06-01    1200.0\n",
      "2008-07-01    1200.0\n",
      "2008-08-01    1200.0\n",
      "2008-09-01    1200.0\n",
      "2008-10-01    1200.0\n",
      "2008-11-01    1200.0\n",
      "2008-12-01    1200.0\n",
      "2009-01-01    1200.0\n",
      "2009-02-01    1200.0\n",
      "2009-03-01    1200.0\n",
      "2009-04-01    1200.0\n",
      "2009-05-01    1200.0\n",
      "2009-06-01    1200.0\n",
      "               ...  \n",
      "2014-07-01    1200.0\n",
      "2014-08-01    1200.0\n",
      "2014-09-01    1200.0\n",
      "2014-10-01    1200.0\n",
      "2014-11-01    1200.0\n",
      "2014-12-01    1200.0\n",
      "2015-01-01    1200.0\n",
      "2015-02-01    1200.0\n",
      "2015-03-01    1200.0\n",
      "2015-04-01    1200.0\n",
      "2015-05-01    1200.0\n",
      "2015-06-01    1200.0\n",
      "2015-07-01    1200.0\n",
      "2015-08-01    1200.0\n",
      "2015-09-01    1200.0\n",
      "2015-10-01    1200.0\n",
      "2015-11-01    1200.0\n",
      "2015-12-01    1200.0\n",
      "2016-01-01    1200.0\n",
      "2016-02-01    1200.0\n",
      "2016-03-01    1200.0\n",
      "2016-04-01    1200.0\n",
      "2016-05-01    1200.0\n",
      "2016-06-01    1200.0\n",
      "2016-07-01    1200.0\n",
      "2016-08-01    1200.0\n",
      "2016-09-01    1200.0\n",
      "2016-10-01    1200.0\n",
      "2016-11-01    1200.0\n",
      "2016-12-01    1200.0\n",
      "Name: Sales, Length: 120, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(df['Sales'].diff(12))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27d0be9550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "np.random.seed(5968)\n",
    "noise=[np.random.normal(scale=500)]\n",
    "\n",
    "for i in range(len(df)-1):\n",
    "    noise.append(np.random.normal(scale=500)+noise[i]*(-0.85))\n",
    "df['Sales2']=df['Sales']+noise\n",
    "df['Sales2'].plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f27e711d1d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(12,8))\n",
    "ax1 = fig.add_subplot(211)\n",
    "fig = sm.graphics.tsa.plot_acf(df['Sales2'].diff().diff(12).dropna(), lags=40, ax=ax1)\n",
    "ax2 = fig.add_subplot(212)\n",
    "fig = sm.graphics.tsa.plot_pacf(df['Sales2'].diff().diff(12).dropna(), lags=40, ax=ax2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(-4.455060575759159, 0.00023686499712739548, 13, 93, {'1%': -3.502704609582561, '5%': -2.8931578098779522, '10%': -2.583636712914788}, 1477.0429266000788)\n"
     ]
    }
   ],
   "source": [
    "print(sm.tsa.stattools.adfuller(df['Sales2'].diff().diff(12).dropna()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "model=sm.tsa.statespace.SARIMAX(endog=df['Sales2'],order=(1,1,0),seasonal_order=(0,1,0,12),trend='c',enforce_invertibility=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "results=model.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                 Statespace Model Results                                 \n",
      "==========================================================================================\n",
      "Dep. Variable:                             Sales2   No. Observations:                  120\n",
      "Model:             SARIMAX(1, 1, 0)x(0, 1, 0, 12)   Log Likelihood                -886.252\n",
      "Date:                            Wed, 21 Aug 2019   AIC                           1778.505\n",
      "Time:                                    23:32:42   BIC                           1786.867\n",
      "Sample:                                01-01-2007   HQIC                          1781.901\n",
      "                                     - 12-01-2016                                         \n",
      "Covariance Type:                              opg                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "intercept    -48.8899     85.873     -0.569      0.569    -217.198     119.418\n",
      "ar.L1         -0.8991      0.037    -24.025      0.000      -0.972      -0.826\n",
      "sigma2      8.107e+05   1.13e+05      7.165      0.000    5.89e+05    1.03e+06\n",
      "===================================================================================\n",
      "Ljung-Box (Q):                      109.04   Jarque-Bera (JB):                14.96\n",
      "Prob(Q):                              0.00   Prob(JB):                         0.00\n",
      "Heteroskedasticity (H):               0.50   Skew:                            -0.47\n",
      "Prob(H) (two-sided):                  0.04   Kurtosis:                         4.57\n",
      "===================================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n"
     ]
    }
   ],
   "source": [
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "model2=sm.tsa.statespace.SARIMAX(endog=df['Sales2'],order=(1,1,0),seasonal_order=(0,1,1,12),trend='c',enforce_invertibility=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "results2=model2.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                 Statespace Model Results                                 \n",
      "==========================================================================================\n",
      "Dep. Variable:                             Sales2   No. Observations:                  120\n",
      "Model:             SARIMAX(1, 1, 0)x(0, 1, 1, 12)   Log Likelihood                -871.263\n",
      "Date:                            Wed, 21 Aug 2019   AIC                           1750.526\n",
      "Time:                                    23:32:44   BIC                           1761.676\n",
      "Sample:                                01-01-2007   HQIC                          1755.054\n",
      "                                     - 12-01-2016                                         \n",
      "Covariance Type:                              opg                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "intercept    -18.8029     28.417     -0.662      0.508     -74.499      36.894\n",
      "ar.L1         -0.9217      0.033    -27.741      0.000      -0.987      -0.857\n",
      "ma.S.L12      -0.6904      0.086     -8.026      0.000      -0.859      -0.522\n",
      "sigma2      4.893e+05   6.45e+04      7.584      0.000    3.63e+05    6.16e+05\n",
      "===================================================================================\n",
      "Ljung-Box (Q):                       51.29   Jarque-Bera (JB):                25.27\n",
      "Prob(Q):                              0.11   Prob(JB):                         0.00\n",
      "Heteroskedasticity (H):               0.53   Skew:                            -0.71\n",
      "Prob(H) (two-sided):                  0.06   Kurtosis:                         4.90\n",
      "===================================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Covariance matrix calculated using the outer product of gradients (complex-step).\n"
     ]
    }
   ],
   "source": [
    "print(results2.summary())"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}