Arima_ann.ipynb

{
 "cells": [
  {
   "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",
    "%matplotlib inline\n",
    "from statsmodels.tsa.arima_model import ARIMA\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv(\"cleaned_Data.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.set_index(\"date\" , inplace = True)\n",
    "data.index = pd.DatetimeIndex(data.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>temp</th>\n",
       "      <th>dew_pt</th>\n",
       "      <th>hum</th>\n",
       "      <th>wind_spd</th>\n",
       "      <th>vis</th>\n",
       "      <th>pressure</th>\n",
       "      <th>rain</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",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-01 02:30:00</th>\n",
       "      <td>69</td>\n",
       "      <td>64</td>\n",
       "      <td>80</td>\n",
       "      <td>2.3</td>\n",
       "      <td>1</td>\n",
       "      <td>29.87</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-01 05:30:00</th>\n",
       "      <td>68</td>\n",
       "      <td>65</td>\n",
       "      <td>87</td>\n",
       "      <td>4.6</td>\n",
       "      <td>1</td>\n",
       "      <td>29.84</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-01 08:30:00</th>\n",
       "      <td>71</td>\n",
       "      <td>64</td>\n",
       "      <td>71</td>\n",
       "      <td>6.9</td>\n",
       "      <td>1</td>\n",
       "      <td>29.92</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-01 11:30:00</th>\n",
       "      <td>79</td>\n",
       "      <td>69</td>\n",
       "      <td>64</td>\n",
       "      <td>4.6</td>\n",
       "      <td>2</td>\n",
       "      <td>29.89</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-01 14:30:00</th>\n",
       "      <td>79</td>\n",
       "      <td>64</td>\n",
       "      <td>51</td>\n",
       "      <td>6.9</td>\n",
       "      <td>2</td>\n",
       "      <td>29.82</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     temp  dew_pt  hum  wind_spd  vis  pressure  rain\n",
       "date                                                                 \n",
       "2018-01-01 02:30:00    69      64   80       2.3    1     29.87     0\n",
       "2018-01-01 05:30:00    68      65   87       4.6    1     29.84     0\n",
       "2018-01-01 08:30:00    71      64   71       6.9    1     29.92     0\n",
       "2018-01-01 11:30:00    79      69   64       4.6    2     29.89     0\n",
       "2018-01-01 14:30:00    79      64   51       6.9    2     29.82     0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "aggregate_data = data.pivot_table(index = data.index.date)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>dew_pt</th>\n",
       "      <th>hum</th>\n",
       "      <th>pressure</th>\n",
       "      <th>rain</th>\n",
       "      <th>temp</th>\n",
       "      <th>vis</th>\n",
       "      <th>wind_spd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-01</th>\n",
       "      <td>65.250000</td>\n",
       "      <td>69.250000</td>\n",
       "      <td>29.861250</td>\n",
       "      <td>0.0</td>\n",
       "      <td>73.375000</td>\n",
       "      <td>1.625000</td>\n",
       "      <td>5.175000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-02</th>\n",
       "      <td>62.500000</td>\n",
       "      <td>68.125000</td>\n",
       "      <td>29.867500</td>\n",
       "      <td>0.0</td>\n",
       "      <td>71.625000</td>\n",
       "      <td>1.625000</td>\n",
       "      <td>4.187500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-03</th>\n",
       "      <td>60.750000</td>\n",
       "      <td>68.250000</td>\n",
       "      <td>29.903750</td>\n",
       "      <td>0.0</td>\n",
       "      <td>69.750000</td>\n",
       "      <td>1.625000</td>\n",
       "      <td>3.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-04</th>\n",
       "      <td>65.200000</td>\n",
       "      <td>69.400000</td>\n",
       "      <td>29.936000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>73.400000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.460000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-05</th>\n",
       "      <td>62.428571</td>\n",
       "      <td>73.285714</td>\n",
       "      <td>29.912857</td>\n",
       "      <td>0.0</td>\n",
       "      <td>69.714286</td>\n",
       "      <td>1.714286</td>\n",
       "      <td>2.957143</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               dew_pt        hum   pressure  rain       temp       vis  \\\n",
       "2018-01-01  65.250000  69.250000  29.861250   0.0  73.375000  1.625000   \n",
       "2018-01-02  62.500000  68.125000  29.867500   0.0  71.625000  1.625000   \n",
       "2018-01-03  60.750000  68.250000  29.903750   0.0  69.750000  1.625000   \n",
       "2018-01-04  65.200000  69.400000  29.936000   0.0  73.400000  2.000000   \n",
       "2018-01-05  62.428571  73.285714  29.912857   0.0  69.714286  1.714286   \n",
       "\n",
       "            wind_spd  \n",
       "2018-01-01  5.175000  \n",
       "2018-01-02  4.187500  \n",
       "2018-01-03  3.600000  \n",
       "2018-01-04  3.460000  \n",
       "2018-01-05  2.957143  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggregate_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "aggregate_data.drop([\"rain\"] , axis=1 , inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[73.375,\n",
       " 71.625,\n",
       " 69.75,\n",
       " 73.4,\n",
       " 69.71428571428571,\n",
       " 69.0,\n",
       " 68.75,\n",
       " 69.0,\n",
       " 67.125,\n",
       " 68.5,\n",
       " 70.0,\n",
       " 72.28571428571429,\n",
       " 72.75,\n",
       " 72.57142857142857,\n",
       " 71.25,\n",
       " 71.125,\n",
       " 71.5,\n",
       " 69.875,\n",
       " 69.75,\n",
       " 69.875,\n",
       " 70.5,\n",
       " 71.25,\n",
       " 71.625,\n",
       " 71.875,\n",
       " 72.75,\n",
       " 72.375,\n",
       " 73.0,\n",
       " 73.71428571428571,\n",
       " 71.375,\n",
       " 72.375,\n",
       " 73.25,\n",
       " 74.375,\n",
       " 74.25,\n",
       " 73.125,\n",
       " 72.875,\n",
       " 73.875,\n",
       " 73.875,\n",
       " 75.75,\n",
       " 76.875,\n",
       " 76.0,\n",
       " 77.85714285714286,\n",
       " 76.75,\n",
       " 76.75,\n",
       " 74.625,\n",
       " 76.25,\n",
       " 76.875,\n",
       " 76.75,\n",
       " 75.125,\n",
       " 76.125,\n",
       " 76.0,\n",
       " 75.875,\n",
       " 76.5,\n",
       " 77.0,\n",
       " 77.0,\n",
       " 77.375,\n",
       " 78.25,\n",
       " 79.375,\n",
       " 79.0,\n",
       " 79.25,\n",
       " 79.5,\n",
       " 79.57142857142857,\n",
       " 80.25,\n",
       " 80.875,\n",
       " 80.375,\n",
       " 80.375,\n",
       " 81.0,\n",
       " 81.75,\n",
       " 81.0,\n",
       " 82.125,\n",
       " 81.25,\n",
       " 80.875,\n",
       " 82.5,\n",
       " 82.0,\n",
       " 80.125,\n",
       " 82.5,\n",
       " 79.14285714285714,\n",
       " 81.0,\n",
       " 81.25,\n",
       " 81.625,\n",
       " 81.375,\n",
       " 81.5,\n",
       " 82.5,\n",
       " 82.5,\n",
       " 82.25,\n",
       " 82.5,\n",
       " 81.75,\n",
       " 80.875,\n",
       " 81.75,\n",
       " 82.0,\n",
       " 81.5,\n",
       " 80.625,\n",
       " 81.83333333333333,\n",
       " 82.125,\n",
       " 81.375,\n",
       " 82.14285714285714,\n",
       " 82.75,\n",
       " 81.25,\n",
       " 81.125,\n",
       " 82.875,\n",
       " 82.125,\n",
       " 82.625,\n",
       " 81.875,\n",
       " 82.25,\n",
       " 82.71428571428571,\n",
       " 84.75,\n",
       " 85.25,\n",
       " 85.0,\n",
       " 83.28571428571429,\n",
       " 84.375,\n",
       " 84.25,\n",
       " 83.625,\n",
       " 83.125,\n",
       " 82.42857142857143,\n",
       " 80.71428571428571,\n",
       " 82.0,\n",
       " 83.57142857142857,\n",
       " 82.83333333333333,\n",
       " 83.0,\n",
       " 82.125,\n",
       " 82.625,\n",
       " 78.25,\n",
       " 82.0,\n",
       " 81.0,\n",
       " 82.875,\n",
       " 83.5,\n",
       " 83.75,\n",
       " 83.25,\n",
       " 82.57142857142857,\n",
       " 81.25,\n",
       " 83.0,\n",
       " 81.875,\n",
       " 82.75,\n",
       " 82.75,\n",
       " 83.375,\n",
       " 84.5,\n",
       " 85.0,\n",
       " 84.75,\n",
       " 85.125,\n",
       " 84.75,\n",
       " 85.125,\n",
       " 84.875,\n",
       " 85.375,\n",
       " 84.0,\n",
       " 85.625,\n",
       " 85.625,\n",
       " 85.625,\n",
       " 86.625,\n",
       " 85.75,\n",
       " 85.375,\n",
       " 85.875,\n",
       " 85.875,\n",
       " 86.25,\n",
       " 86.0,\n",
       " 85.875,\n",
       " 87.0,\n",
       " 88.5,\n",
       " 87.5,\n",
       " 87.0,\n",
       " 84.5,\n",
       " 84.5,\n",
       " 86.25,\n",
       " 86.625,\n",
       " 86.0,\n",
       " 85.57142857142857,\n",
       " 85.625,\n",
       " 85.875,\n",
       " 86.0,\n",
       " 86.75,\n",
       " 87.75,\n",
       " 87.5,\n",
       " 86.625,\n",
       " 84.125,\n",
       " 83.75,\n",
       " 84.0,\n",
       " 85.75,\n",
       " 84.75,\n",
       " 84.375,\n",
       " 79.5,\n",
       " 80.14285714285714,\n",
       " 81.0,\n",
       " 83.4,\n",
       " 84.0,\n",
       " 84.0,\n",
       " 84.625,\n",
       " 86.75,\n",
       " 84.875,\n",
       " 81.875,\n",
       " 83.25,\n",
       " 84.625,\n",
       " 83.875,\n",
       " 83.75,\n",
       " 84.0,\n",
       " 83.625,\n",
       " 81.0,\n",
       " 80.25,\n",
       " 80.75,\n",
       " 83.625,\n",
       " 85.125,\n",
       " 85.875,\n",
       " 79.5,\n",
       " 77.625,\n",
       " 81.85714285714286,\n",
       " 84.5,\n",
       " 84.25,\n",
       " 84.875,\n",
       " 85.25,\n",
       " 85.875,\n",
       " 86.25,\n",
       " 86.125,\n",
       " 84.0,\n",
       " 83.25,\n",
       " 83.625,\n",
       " 82.125,\n",
       " 81.75,\n",
       " 86.375,\n",
       " 87.28571428571429,\n",
       " 85.0,\n",
       " 80.0,\n",
       " 83.85714285714286,\n",
       " 84.0,\n",
       " 84.875,\n",
       " 86.75,\n",
       " 84.125,\n",
       " 82.25,\n",
       " 81.625,\n",
       " 82.625,\n",
       " 84.0,\n",
       " 84.625,\n",
       " 84.75,\n",
       " 83.5,\n",
       " 82.5,\n",
       " 85.625,\n",
       " 87.125,\n",
       " 86.75,\n",
       " 85.625,\n",
       " 82.875,\n",
       " 77.75,\n",
       " 81.0,\n",
       " 83.5,\n",
       " 82.75,\n",
       " 83.75,\n",
       " 83.375,\n",
       " 82.125,\n",
       " 83.75,\n",
       " 84.75,\n",
       " 84.875,\n",
       " 83.125,\n",
       " 79.25,\n",
       " 82.71428571428571,\n",
       " 84.25,\n",
       " 85.875,\n",
       " 85.25,\n",
       " 84.25,\n",
       " 84.875,\n",
       " 86.125,\n",
       " 86.125,\n",
       " 87.0,\n",
       " 85.42857142857143,\n",
       " 84.375,\n",
       " 85.5,\n",
       " 85.625,\n",
       " 79.25,\n",
       " 81.875,\n",
       " 84.125,\n",
       " 84.375,\n",
       " 84.75,\n",
       " 84.25,\n",
       " 84.5,\n",
       " 84.75,\n",
       " 85.375,\n",
       " 85.875,\n",
       " 85.75,\n",
       " 87.28571428571429,\n",
       " 86.0,\n",
       " 87.125,\n",
       " 86.0,\n",
       " 85.375,\n",
       " 85.375,\n",
       " 87.85714285714286,\n",
       " 86.625,\n",
       " 83.875,\n",
       " 76.625,\n",
       " 81.125,\n",
       " 82.5,\n",
       " 82.75,\n",
       " 83.5,\n",
       " 83.375,\n",
       " 83.125,\n",
       " 81.75,\n",
       " 81.5,\n",
       " 83.375,\n",
       " 83.625,\n",
       " 84.125,\n",
       " 83.25,\n",
       " 82.875,\n",
       " 82.875,\n",
       " 81.75,\n",
       " 79.25,\n",
       " 77.75,\n",
       " 75.125,\n",
       " 78.75,\n",
       " 78.25,\n",
       " 79.125,\n",
       " 81.0]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggregate_data['temp'].values.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "column_list = aggregate_data[\"temp\"].values.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_len = int(len(column_list)*0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "train,test = column_list[:training_len] , column_list[training_len:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_data = train[:]\n",
    "error_list = []\n",
    "prediction_list = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\signal\\signaltools.py:1341: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  out_full[ind] += zi\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\signal\\signaltools.py:1344: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  out = out_full[ind]\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\signal\\signaltools.py:1350: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  zf = out_full[ind]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "predicted=84.238777, expected=82.125000 error=2.113777\n",
      "predicted=82.643313, expected=81.750000 error=0.893313\n",
      "predicted=82.338131, expected=86.375000 error=4.036869\n",
      "predicted=85.951965, expected=87.285714 error=1.333749\n",
      "predicted=86.044725, expected=85.000000 error=1.044725\n",
      "predicted=84.132735, expected=80.000000 error=4.132735\n",
      "predicted=80.714099, expected=83.857143 error=3.143044\n",
      "predicted=84.295188, expected=84.000000 error=0.295188\n",
      "predicted=84.469969, expected=84.875000 error=0.405031\n",
      "predicted=84.078146, expected=86.750000 error=2.671854\n",
      "predicted=86.634817, expected=84.125000 error=2.509817\n",
      "predicted=82.922367, expected=82.250000 error=0.672367\n",
      "predicted=82.798271, expected=81.625000 error=1.173271\n",
      "predicted=82.693507, expected=82.625000 error=0.068507\n",
      "predicted=83.126377, expected=84.000000 error=0.873623\n",
      "predicted=84.331887, expected=84.625000 error=0.293113\n",
      "predicted=84.060778, expected=84.750000 error=0.689222\n",
      "predicted=83.978804, expected=83.500000 error=0.478804\n",
      "predicted=83.071315, expected=82.500000 error=0.571315\n",
      "predicted=82.652171, expected=85.625000 error=2.972829\n",
      "predicted=85.521370, expected=87.125000 error=1.603630\n",
      "predicted=86.291208, expected=86.750000 error=0.458792\n",
      "predicted=85.644657, expected=85.625000 error=0.019657\n",
      "predicted=85.144003, expected=82.875000 error=2.269003\n",
      "predicted=83.005111, expected=77.750000 error=5.255111\n",
      "predicted=79.621491, expected=81.000000 error=1.378509\n",
      "predicted=83.135657, expected=83.500000 error=0.364343\n",
      "predicted=84.319526, expected=82.750000 error=1.569526\n",
      "predicted=82.250065, expected=83.750000 error=1.499935\n",
      "predicted=83.359883, expected=83.375000 error=0.015117\n",
      "predicted=81.918460, expected=82.125000 error=0.206540\n",
      "predicted=81.898046, expected=83.750000 error=1.851954\n",
      "predicted=84.047773, expected=84.750000 error=0.702227\n",
      "predicted=84.172129, expected=84.875000 error=0.702871\n",
      "predicted=84.334743, expected=83.125000 error=1.209743\n",
      "predicted=83.010948, expected=79.250000 error=3.760948\n",
      "predicted=79.819946, expected=82.714286 error=2.894340\n",
      "predicted=83.645965, expected=84.250000 error=0.604035\n",
      "predicted=84.320216, expected=85.875000 error=1.554784\n",
      "predicted=84.870550, expected=85.250000 error=0.379450\n",
      "predicted=84.302732, expected=84.250000 error=0.052732\n",
      "predicted=82.887002, expected=84.875000 error=1.987998\n",
      "predicted=84.838409, expected=86.125000 error=1.286591\n",
      "predicted=85.856771, expected=86.125000 error=0.268229\n",
      "predicted=85.829959, expected=87.000000 error=1.170041\n",
      "predicted=86.382758, expected=85.428571 error=0.954187\n",
      "predicted=84.999798, expected=84.375000 error=0.624798\n",
      "predicted=84.543973, expected=85.500000 error=0.956027\n",
      "predicted=85.966598, expected=85.625000 error=0.341598\n",
      "predicted=85.613434, expected=79.250000 error=6.363434\n",
      "predicted=80.478672, expected=81.875000 error=1.396328\n",
      "predicted=83.186248, expected=84.125000 error=0.938752\n",
      "predicted=84.352815, expected=84.375000 error=0.022185\n",
      "predicted=83.760842, expected=84.750000 error=0.989158\n",
      "predicted=84.575170, expected=84.250000 error=0.325170\n",
      "predicted=82.773923, expected=84.500000 error=1.726077\n",
      "predicted=84.153294, expected=84.750000 error=0.596706\n",
      "predicted=84.723725, expected=85.375000 error=0.651275\n",
      "predicted=85.046738, expected=85.875000 error=0.828262\n",
      "predicted=85.480976, expected=85.750000 error=0.269024\n",
      "predicted=85.229779, expected=87.285714 error=2.055935\n",
      "predicted=86.662396, expected=86.000000 error=0.662396\n",
      "predicted=85.541925, expected=87.125000 error=1.583075\n",
      "predicted=86.744188, expected=86.000000 error=0.744188\n",
      "predicted=85.973605, expected=85.375000 error=0.598605\n",
      "predicted=85.389804, expected=85.375000 error=0.014804\n",
      "predicted=85.948006, expected=87.857143 error=1.909137\n",
      "predicted=87.349538, expected=86.625000 error=0.724538\n",
      "predicted=86.352903, expected=83.875000 error=2.477903\n",
      "predicted=84.046688, expected=76.625000 error=7.421688\n",
      "predicted=78.618819, expected=81.125000 error=2.506181\n",
      "predicted=83.189099, expected=82.500000 error=0.689099\n",
      "predicted=83.846784, expected=82.750000 error=1.096784\n",
      "predicted=82.116597, expected=83.500000 error=1.383403\n",
      "predicted=83.110084, expected=83.375000 error=0.264916\n",
      "predicted=81.585976, expected=83.125000 error=1.539024\n",
      "predicted=82.913256, expected=81.750000 error=1.163256\n",
      "predicted=81.998969, expected=81.500000 error=0.498969\n",
      "predicted=81.927826, expected=83.375000 error=1.447174\n",
      "predicted=83.536777, expected=83.625000 error=0.088223\n",
      "predicted=83.230794, expected=84.125000 error=0.894206\n",
      "predicted=83.539695, expected=83.250000 error=0.289695\n",
      "predicted=82.817587, expected=82.875000 error=0.057413\n",
      "predicted=82.663706, expected=82.875000 error=0.211294\n",
      "predicted=83.168607, expected=81.750000 error=1.418607\n",
      "predicted=82.108650, expected=79.250000 error=2.858650\n",
      "predicted=80.284202, expected=77.750000 error=2.534202\n",
      "predicted=79.156433, expected=75.125000 error=4.031433\n",
      "predicted=76.865012, expected=78.750000 error=1.884988\n",
      "predicted=79.853656, expected=78.250000 error=1.603656\n",
      "predicted=78.576001, expected=79.125000 error=0.548999\n",
      "predicted=78.495385, expected=81.000000 error=2.504615\n"
     ]
    }
   ],
   "source": [
    "for i in test:\n",
    "    model = ARIMA(training_data , order=(7,0,0))\n",
    "    model_fit = model.fit(disp=0)\n",
    "    predicted_value = model_fit.forecast()\n",
    "    predicted_value = predicted_value[0]\n",
    "    error = abs(predicted_value-i)\n",
    "    prediction_list.append(predicted_value)\n",
    "    error_list.append(error)\n",
    "    training_data.append(i)\n",
    "    print('predicted=%f, expected=%f error=%f' % (predicted_value , i ,error))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x25c1a7ad940>]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(prediction_list,color='r')\n",
    "plt.plot(test,color='g')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "error_list = np.array(error_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.7563689999117145"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(test ,prediction_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\signal\\signaltools.py:1341: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  out_full[ind] += zi\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\signal\\signaltools.py:1344: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  out = out_full[ind]\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\scipy\\signal\\signaltools.py:1350: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  zf = out_full[ind]\n"
     ]
    }
   ],
   "source": [
    "model = ARIMA(train,order=(7,0,0))\n",
    "model_fit_train = model.fit(disp=0)\n",
    "training_error = model_fit_train.resid\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-6.00152139, -1.92871942, -2.25285752,  2.97603882, -3.98779573,\n",
       "       -1.43057506, -0.73565574, -1.03300871, -1.86707545,  0.05393363,\n",
       "        1.30566709,  2.21246566,  1.37378555,  0.53392466, -0.2924325 ,\n",
       "       -0.05239185,  0.19403685, -2.13341725, -0.647965  , -0.48970215,\n",
       "        0.11244672,  0.66058329,  0.28411495,  0.61844949,  1.17347773,\n",
       "       -0.0157262 ,  0.82979505,  0.92508688, -2.17906678,  0.90128525,\n",
       "        0.57294313,  0.92953735,  0.36066365, -1.04497407,  0.09282788,\n",
       "        0.74551315, -0.16020965,  1.78105651,  1.49017312, -0.20463986,\n",
       "        2.58870455, -0.59963316,  0.3079057 , -1.89071822,  0.83415203,\n",
       "        0.63354104, -0.54987394, -1.20133051,  0.60802486,  0.01464812,\n",
       "       -0.42689204,  0.62471817,  0.34229539,  0.38321051,  0.54185887,\n",
       "        1.08498837,  1.48171105,  0.15763089,  0.67215592,  0.64393021,\n",
       "        0.26622849,  0.91220314,  0.74202529, -0.12422708,  0.28367843,\n",
       "        0.75185448,  0.88955103, -0.42902184,  1.24856977, -0.52420013,\n",
       "       -0.30632059,  1.72265472, -0.56628079, -1.52627988,  2.1810498 ,\n",
       "       -3.33896983,  1.50640595,  0.20444433, -0.14827123,  0.6167797 ,\n",
       "       -0.22826473,  1.58005143,  0.1432272 ,  0.0741837 ,  0.47706418,\n",
       "       -0.607647  , -0.83977626,  0.57702165,  0.01623379, -0.47488538,\n",
       "       -0.82712159,  1.00986494,  0.40238737, -0.68334152,  0.91001372,\n",
       "        0.68255989, -1.13324391, -0.08843323,  1.52144257, -0.63769041,\n",
       "        0.69675181, -0.56629122,  0.38043157,  0.75675544,  1.87077502,\n",
       "        1.19266082,  0.38721371, -0.86997632,  1.14025438,  0.00903587,\n",
       "       -0.86305216, -0.46873447, -1.06238403, -1.82690284,  0.59244984,\n",
       "        1.13429621, -0.69327529,  0.58131272, -0.6442451 ,  0.62801815,\n",
       "       -4.27683386,  2.66487813, -1.14413546,  1.23914541,  1.602003  ,\n",
       "        0.15516404,  0.9861649 , -0.68786053, -1.04858635,  1.27468727,\n",
       "       -1.28410244,  0.596089  ,  0.25866451,  0.52669902,  1.78104531,\n",
       "        0.7097136 ,  0.53071251,  0.83249906,  0.03699257,  0.57642004,\n",
       "       -0.08496867,  0.47966326, -1.11222713,  1.44802357,  0.26328096,\n",
       "        0.04863205,  1.52319286, -0.7070552 ,  0.10557528,  0.5191555 ,\n",
       "       -0.06770263,  0.55904254, -0.16196744,  0.03898767,  1.34166535,\n",
       "        1.70957942, -0.3587236 ,  0.01966125, -2.20011649, -0.45192549,\n",
       "        1.22407835, -0.18547866, -0.28418879, -0.22420077,  0.30363959,\n",
       "        0.37521453,  0.07023593,  0.78989321,  1.30552102,  0.29363513,\n",
       "       -0.32812901, -2.26112068, -0.81586641, -0.45293734,  0.92684459,\n",
       "       -0.96590513, -0.37427626, -4.50710685, -0.36276701, -0.02529881,\n",
       "        1.09025661,  1.06054458,  0.270041  ,  1.91202162,  2.7544127 ,\n",
       "       -0.90160583, -2.64011539,  1.08486493,  0.74576532, -0.82120232,\n",
       "       -0.13312043,  0.26010978, -0.02639039, -2.53133735, -1.38886937,\n",
       "       -0.25853874,  2.10479478,  1.68774647,  1.42981977, -4.95979056,\n",
       "       -2.05497244,  3.20882281,  1.3610739 ,  0.23604771,  1.25944111,\n",
       "        1.62205792,  1.78566759,  0.98585062,  0.14964535, -1.65673407,\n",
       "       -0.90194033, -0.0980596 ])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x = []\n",
    "train_y = []\n",
    "for i in range(len(training_error)-7):\n",
    "    train_x.append(training_error[i:i+7])\n",
    "    train_y.append(training_error[i+7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1.0330087052133923"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_y[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-6.00152139, -1.92871942, -2.25285752,  2.97603882, -3.98779573,\n",
       "       -1.43057506, -0.73565574])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_x = []\n",
    "test_y = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(len(error_list)-7):\n",
    "    test_x.append(error_list[i:i+7])\n",
    "    test_y.append(error_list[i+7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_train_x = []\n",
    "new_train_y = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in train_x:\n",
    "    new_train_x.append(i.reshape(-1))\n",
    "    \n",
    "for i in train_y:\n",
    "    new_train_y.append(i.reshape(-1))\n",
    "new_train_x = np.array(new_train_x)\n",
    "new_train_y = np.array(new_train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense,Activation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:1: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(100, activation=\"tanh\", input_dim=7, kernel_initializer=\"uniform\")`\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:2: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(50, activation=\"tanh\", kernel_initializer=\"uniform\")`\n",
      "  \n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:3: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(25, activation=\"tanh\", kernel_initializer=\"uniform\")`\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    }
   ],
   "source": [
    "model.add(Dense(100 , activation=\"tanh\" , init=\"uniform\" , input_dim=7))\n",
    "model.add(Dense(50 , activation=\"tanh\" , init=\"uniform\"))\n",
    "model.add(Dense(25 , activation=\"tanh\" , init=\"uniform\"))\n",
    "model.add(Dense(1 , activation=\"linear\"))\n",
    "model.compile(\"Adam\", loss=\"mean_squared_error\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x25c1f789f98>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(new_train_x , new_train_y , batch_size=20, epochs=400,verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_test_x = []\n",
    "new_test_y = []\n",
    "for i in test_x:\n",
    "    new_test_x.append(i.reshape(-1))\n",
    "    \n",
    "for i in test_y:\n",
    "    new_test_y.append(i.reshape(-1))\n",
    "new_test_x = np.array(new_test_x)\n",
    "new_test_y = np.array(new_test_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_error = model.predict(new_test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x25c20d1e5c0>]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(test_y)\n",
    "plt.plot(predict_error, color=\"r\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "arima_ann_predict = prediction_list[7:]+predict_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x25c20d52668>]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(arima_ann_predict,color='r')\n",
    "plt.plot(test[7:],color='g')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}