Xperty.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Stock Prices predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DER_EU_ENXT_ALL_SUM_20190520.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "import os\n",
    "from subprocess import check_output\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "from pandas.plotting import lag_plot\n",
    "from pandas import datetime\n",
    "from statsmodels.tsa.arima_model import ARIMA\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import seaborn as sns\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report,confusion_matrix\n",
    "from itertools import product\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %matplotlib inline\n",
    "# import mpld3\n",
    "# mpld3.enable_notebook()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['minidata', 'level2stockanalysis']\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "print(os.listdir(\"../input\"))"
   ]
  },
  {
   "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>H;DER_EU_ENXT_ALL_SUM.csv;2019-05-21 04:35:30</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>D;MarketPlace;AssetClass;AssetGroup;ExchangeCo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>E;AMS;DRV;EQU;A;F;AA6;AFAA6190600000F;20190621...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>E;AMS;DRV;EQU;A;F;AA6;AFAA6190700000F;20190719...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>E;AMS;DRV;EQU;A;F;AA6;AFAA6190800000F;20190816...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>E;AMS;DRV;EQU;A;F;AA6;AFAA6190900000F;20190920...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       H;DER_EU_ENXT_ALL_SUM.csv;2019-05-21 04:35:30\n",
       "0  D;MarketPlace;AssetClass;AssetGroup;ExchangeCo...\n",
       "1  E;AMS;DRV;EQU;A;F;AA6;AFAA6190600000F;20190621...\n",
       "2  E;AMS;DRV;EQU;A;F;AA6;AFAA6190700000F;20190719...\n",
       "3  E;AMS;DRV;EQU;A;F;AA6;AFAA6190800000F;20190816...\n",
       "4  E;AMS;DRV;EQU;A;F;AA6;AFAA6190900000F;20190920..."
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"../input/level2stockanalysis/DER_EU_ENXT_ALL_SUM_20190520.csv\")\n",
    "df.head()"
   ]
  },
  {
   "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>D</th>\n",
       "      <th>MarketPlace</th>\n",
       "      <th>AssetClass</th>\n",
       "      <th>AssetGroup</th>\n",
       "      <th>ExchangeCode</th>\n",
       "      <th>InstrumentType</th>\n",
       "      <th>InstrumentTrading Symbol</th>\n",
       "      <th>InstrumentID</th>\n",
       "      <th>ExpiryDate</th>\n",
       "      <th>ExercisePrice</th>\n",
       "      <th>ActualExercisePrice</th>\n",
       "      <th>OptionCategory</th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>TradeCount</th>\n",
       "      <th>Close</th>\n",
       "      <th>Settlement</th>\n",
       "      <th>COBVolume</th>\n",
       "      <th>OffCOBVolume</th>\n",
       "      <th>TotalVolume</th>\n",
       "      <th>OpenInterest</th>\n",
       "      <th>OpenInterestDate</th>\n",
       "      <th>EndofDayDelta</th>\n",
       "      <th>Volatility</th>\n",
       "      <th>VolatilityForValuation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>E</td>\n",
       "      <td>AMS</td>\n",
       "      <td>DRV</td>\n",
       "      <td>EQU</td>\n",
       "      <td>A</td>\n",
       "      <td>F</td>\n",
       "      <td>AA6</td>\n",
       "      <td>AFAA6190600000F</td>\n",
       "      <td>20190621</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.7306</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20190520</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>E</td>\n",
       "      <td>AMS</td>\n",
       "      <td>DRV</td>\n",
       "      <td>EQU</td>\n",
       "      <td>A</td>\n",
       "      <td>F</td>\n",
       "      <td>AA6</td>\n",
       "      <td>AFAA6190700000F</td>\n",
       "      <td>20190719</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.7223</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20190520</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>E</td>\n",
       "      <td>AMS</td>\n",
       "      <td>DRV</td>\n",
       "      <td>EQU</td>\n",
       "      <td>A</td>\n",
       "      <td>F</td>\n",
       "      <td>AA6</td>\n",
       "      <td>AFAA6190800000F</td>\n",
       "      <td>20190816</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.7141</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20190520</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>E</td>\n",
       "      <td>AMS</td>\n",
       "      <td>DRV</td>\n",
       "      <td>EQU</td>\n",
       "      <td>A</td>\n",
       "      <td>F</td>\n",
       "      <td>AA6</td>\n",
       "      <td>AFAA6190900000F</td>\n",
       "      <td>20190920</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.7038</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20190520</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>E</td>\n",
       "      <td>AMS</td>\n",
       "      <td>DRV</td>\n",
       "      <td>EQU</td>\n",
       "      <td>A</td>\n",
       "      <td>F</td>\n",
       "      <td>AA6</td>\n",
       "      <td>AFAA6191200000F</td>\n",
       "      <td>20191220</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30.677</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20190520</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "0  D MarketPlace          ...           Volatility VolatilityForValuation\n",
       "1  E         AMS          ...                    0               0.000000\n",
       "2  E         AMS          ...                    0               0.000000\n",
       "3  E         AMS          ...                    0               0.000000\n",
       "4  E         AMS          ...                    0               0.000000\n",
       "5  E         AMS          ...                    0               0.000000\n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.iloc[:,0].str.split(';', expand=True)\n",
    "\n",
    "new_header = df.iloc[0] #grab the first row for the header\n",
    "df = df[1:] #take the data less the header row\n",
    "df.columns = new_header #set the header row as the df header\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100894, 26)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['D', 'MarketPlace', 'AssetClass', 'AssetGroup', 'ExchangeCode',\n",
      "       'InstrumentType', 'InstrumentTrading Symbol', 'InstrumentID',\n",
      "       'ExpiryDate', 'ExercisePrice', 'ActualExercisePrice', 'OptionCategory',\n",
      "       'Open', 'High', 'Low', 'TradeCount', 'Close', 'Settlement', 'COBVolume',\n",
      "       'OffCOBVolume', 'TotalVolume', 'OpenInterest', 'OpenInterestDate',\n",
      "       'EndofDayDelta', 'Volatility', 'VolatilityForValuation'],\n",
      "      dtype='object', name=0)\n"
     ]
    }
   ],
   "source": [
    "print(df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>percent_missing</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>D</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MarketPlace</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AssetClass</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AssetGroup</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExchangeCode</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InstrumentType</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InstrumentTrading Symbol</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InstrumentID</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExpiryDate</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExercisePrice</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ActualExercisePrice</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OptionCategory</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Open</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>High</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Low</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TradeCount</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Close</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Settlement</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COBVolume</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OffCOBVolume</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TotalVolume</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OpenInterest</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OpenInterestDate</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EndofDayDelta</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Volatility</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VolatilityForValuation</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          percent_missing\n",
       "0                                        \n",
       "D                                0.000000\n",
       "MarketPlace                      0.000000\n",
       "AssetClass                       0.000000\n",
       "AssetGroup                       0.000991\n",
       "ExchangeCode                     0.000991\n",
       "InstrumentType                   0.000991\n",
       "InstrumentTrading Symbol         0.000991\n",
       "InstrumentID                     0.000991\n",
       "ExpiryDate                       0.000991\n",
       "ExercisePrice                    0.000991\n",
       "ActualExercisePrice              0.000991\n",
       "OptionCategory                   0.000991\n",
       "Open                             0.000991\n",
       "High                             0.000991\n",
       "Low                              0.000991\n",
       "TradeCount                       0.000991\n",
       "Close                            0.000991\n",
       "Settlement                       0.000991\n",
       "COBVolume                        0.000991\n",
       "OffCOBVolume                     0.000991\n",
       "TotalVolume                      0.000991\n",
       "OpenInterest                     0.000991\n",
       "OpenInterestDate                 0.000991\n",
       "EndofDayDelta                    0.000991\n",
       "Volatility                       0.000991\n",
       "VolatilityForValuation           0.000991"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percent_missing = df.isnull().sum() * 100 / len(df)\n",
    "missing_values = pd.DataFrame({'percent_missing': percent_missing})\n",
    "missing_values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Settlement'] = pd.to_numeric(df['Settlement'], errors='coerce')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Settlement'].plot()\n",
    "plt.title(\"Settlement\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['VolatilityForValuation'] = pd.to_numeric(df['VolatilityForValuation'], errors='coerce')\n",
    "df['VolatilityForValuation'].plot()\n",
    "plt.title(\"VolatilityForValuation\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['ExercisePrice'] = pd.to_numeric(df['ExercisePrice'], errors='coerce')\n",
    "#df['ExercisePrice'].value_counts()\n",
    "df['ExercisePrice'].plot()\n",
    "plt.title(\"ExercisePrice\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Low'] = pd.to_numeric(df['Low'], errors='coerce')\n",
    "#df['ExercisePrice'].value_counts()\n",
    "df['Low'].plot()\n",
    "plt.title(\"Low\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['High'] = pd.to_numeric(df['High'], errors='coerce')\n",
    "#df['ExercisePrice'].value_counts()\n",
    "df['High'].plot()\n",
    "plt.title(\"High\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Open'] = pd.to_numeric(df['Open'], errors='coerce')\n",
    "#df['ExercisePrice'].value_counts()\n",
    "df['Open'].plot()\n",
    "plt.title(\"Open\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Close'] = pd.to_numeric(df['Close'], errors='coerce')\n",
    "#df['ExercisePrice'].value_counts()\n",
    "df['Close'].plot()\n",
    "plt.title(\"Close\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Machine Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[['ExercisePrice', 'ActualExercisePrice',\n",
    "       'Open', 'High', 'Low', 'TradeCount', 'Close', 'Settlement', 'COBVolume',\n",
    "       'OffCOBVolume', 'TotalVolume',\n",
    "       'EndofDayDelta', 'Volatility']]\n",
    "X = X.apply(pd.to_numeric, errors='coerce')\n",
    "X = X.fillna(0)\n",
    "\n",
    "Y = df['VolatilityForValuation'].fillna(0)\n",
    "\n",
    "X = StandardScaler().fit_transform(X)\n",
    "\n",
    "X_Train, X_Test, Y_Train, Y_Test = train_test_split(X, Y, test_size = 0.05, random_state = 101)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def smape_kun(y_true, y_pred):\n",
    "    return np.mean((np.abs(y_pred - y_true) * 200/ (np.abs(y_pred) + np.abs(y_true))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.0322448329100822\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "model = RandomForestRegressor(n_jobs=-1000)\n",
    "\n",
    "model.fit(X_Train,Y_Train)\n",
    "\n",
    "y_pred = model.predict(X_Test)\n",
    "\n",
    "print(smape_kun(Y_Test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "79.94974981120832\n"
     ]
    }
   ],
   "source": [
    "from sklearn import linear_model\n",
    "clf = linear_model.Lasso(alpha=0.1)\n",
    "\n",
    "clf.fit(X_Train,Y_Train)\n",
    "\n",
    "y_pred = clf.predict(X_Test)\n",
    "\n",
    "print(smape_kun(Y_Test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "80.02601229757758\n",
      "[27.45681216 11.44692291 24.43405989 ... 25.46844104  0.41435491\n",
      "  8.62502859]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "reg = LinearRegression().fit(X_Train,Y_Train)\n",
    "\n",
    "y_pred = reg.predict(X_Test)\n",
    "\n",
    "print(smape_kun(Y_Test,y_pred))\n",
    "print(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.1429487504424054\n"
     ]
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor\n",
    "tree_regressor = DecisionTreeRegressor(random_state = 0)\n",
    "tree_regressor.fit(X_Train,Y_Train)\n",
    "\n",
    "y_pred = tree_regressor.predict(X_Test)\n",
    "\n",
    "print(smape_kun(Y_Test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "multiclass\n"
     ]
    }
   ],
   "source": [
    "from sklearn import preprocessing\n",
    "from sklearn import utils\n",
    "\n",
    "lab_enc = preprocessing.LabelEncoder()\n",
    "Y_Train = lab_enc.fit_transform(Y_Train)\n",
    "print(utils.multiclass.type_of_target(Y_Train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/discriminant_analysis.py:388: UserWarning: Variables are collinear.\n",
      "  warnings.warn(\"Variables are collinear.\")\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "197.44040632212443\n"
     ]
    }
   ],
   "source": [
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "\n",
    "trainedlda = LinearDiscriminantAnalysis().fit(X_Train, Y_Train)\n",
    "predictionlda = trainedlda.predict(X_Test)\n",
    "print(smape_kun(Y_Test,predictionlda))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### https://github.com/susanli2016/Machine-Learning-with-Python/blob/master/Time%20Series%20LSTM%20VIX.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data shape:  (100874, 20)\n",
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:58: UserWarning: The `input_dim` and `input_length` arguments in recurrent layers are deprecated. Use `input_shape` instead.\n",
      "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:58: UserWarning: Update your `LSTM` call to the Keras 2 API: `LSTM(return_sequences=True, input_shape=(None, 1), units=50)`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:61: UserWarning: Update your `LSTM` call to the Keras 2 API: `LSTM(return_sequences=False, units=100)`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:65: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"linear\", units=1)`\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 76664 samples, validate on 4035 samples\n",
      "Epoch 1/100\n",
      "76664/76664 [==============================] - 23s 303us/step - loss: 1107.0870 - val_loss: 8286.3328\n",
      "Epoch 2/100\n",
      "76664/76664 [==============================] - 21s 277us/step - loss: 1105.9221 - val_loss: 8285.5120\n",
      "Epoch 3/100\n",
      "76664/76664 [==============================] - 21s 277us/step - loss: 1105.0541 - val_loss: 8283.3020\n",
      "Epoch 4/100\n",
      "76664/76664 [==============================] - 21s 277us/step - loss: 1104.4473 - val_loss: 8282.9193\n",
      "Epoch 5/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1104.3094 - val_loss: 8281.6626\n",
      "Epoch 6/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1104.1902 - val_loss: 8279.5012\n",
      "Epoch 7/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1103.7844 - val_loss: 8280.2761\n",
      "Epoch 8/100\n",
      "76664/76664 [==============================] - 21s 280us/step - loss: 1103.2841 - val_loss: 8280.0987\n",
      "Epoch 9/100\n",
      "76664/76664 [==============================] - 22s 281us/step - loss: 1103.3341 - val_loss: 8279.5995\n",
      "Epoch 10/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1103.3020 - val_loss: 8276.6158\n",
      "Epoch 11/100\n",
      "76664/76664 [==============================] - 22s 289us/step - loss: 1103.0428 - val_loss: 8277.4442\n",
      "Epoch 12/100\n",
      "76664/76664 [==============================] - 22s 281us/step - loss: 1102.6828 - val_loss: 8278.2615\n",
      "Epoch 13/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1102.3647 - val_loss: 8275.9646\n",
      "Epoch 14/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1102.0798 - val_loss: 8282.7926\n",
      "Epoch 15/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1103.7616 - val_loss: 8278.6516\n",
      "Epoch 16/100\n",
      "76664/76664 [==============================] - 22s 282us/step - loss: 1102.1631 - val_loss: 8280.0777\n",
      "Epoch 17/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1102.5755 - val_loss: 8276.8700\n",
      "Epoch 18/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1102.2885 - val_loss: 8276.4282\n",
      "Epoch 19/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1102.4316 - val_loss: 8275.5922\n",
      "Epoch 20/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1101.9809 - val_loss: 8274.9366\n",
      "Epoch 21/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1102.2120 - val_loss: 8273.5415\n",
      "Epoch 22/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1101.1452 - val_loss: 8277.4853\n",
      "Epoch 23/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1101.6323 - val_loss: 8280.5501\n",
      "Epoch 24/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1102.4754 - val_loss: 8271.9484\n",
      "Epoch 25/100\n",
      "76664/76664 [==============================] - 23s 297us/step - loss: 1101.3341 - val_loss: 8277.1337\n",
      "Epoch 26/100\n",
      "76664/76664 [==============================] - 22s 283us/step - loss: 1101.6066 - val_loss: 8270.3921\n",
      "Epoch 27/100\n",
      "76664/76664 [==============================] - 22s 282us/step - loss: 1100.2038 - val_loss: 8270.6334\n",
      "Epoch 28/100\n",
      "76664/76664 [==============================] - 21s 280us/step - loss: 1101.3755 - val_loss: 8275.2005\n",
      "Epoch 29/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1101.0215 - val_loss: 8277.9650\n",
      "Epoch 30/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1100.5674 - val_loss: 8275.1865\n",
      "Epoch 31/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1101.6544 - val_loss: 8276.4276\n",
      "Epoch 32/100\n",
      "76664/76664 [==============================] - 21s 276us/step - loss: 1100.7251 - val_loss: 8277.3071\n",
      "Epoch 33/100\n",
      "76664/76664 [==============================] - 21s 277us/step - loss: 1098.7112 - val_loss: 8275.4045\n",
      "Epoch 34/100\n",
      "76664/76664 [==============================] - 21s 277us/step - loss: 1100.7585 - val_loss: 8273.7195\n",
      "Epoch 35/100\n",
      "76664/76664 [==============================] - 21s 276us/step - loss: 1099.4730 - val_loss: 8274.9292\n",
      "Epoch 36/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1099.8430 - val_loss: 8272.0571\n",
      "Epoch 37/100\n",
      "76664/76664 [==============================] - 21s 278us/step - loss: 1100.7206 - val_loss: 8271.2664\n",
      "Epoch 38/100\n",
      "76664/76664 [==============================] - 22s 282us/step - loss: 1099.8020 - val_loss: 8274.5095\n",
      "Epoch 39/100\n",
      "76664/76664 [==============================] - 21s 279us/step - loss: 1099.8701 - val_loss: 8273.2615\n",
      "Epoch 40/100\n",
      "22016/76664 [=======>......................] - ETA: 15s - loss: 1317.4790"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import time\n",
    "from keras.layers.core import Dense, Activation, Dropout\n",
    "from keras.layers.recurrent import LSTM\n",
    "from keras.models import Sequential\n",
    "\n",
    "def convertSeriesToMatrix(vectorSeries, sequence_length):\n",
    "    matrix=[]\n",
    "    for i in range(len(vectorSeries)-sequence_length+1):\n",
    "        matrix.append(vectorSeries[i:i+sequence_length])\n",
    "    return matrix\n",
    "\n",
    "np.random.seed(2019)\n",
    "path_to_dataset = \"../input/minidata/mycsvfile.csv\"\n",
    "sequence_length = 20\n",
    "\n",
    "\n",
    "# vector to store the time series\n",
    "vector_vix = []\n",
    "with open(path_to_dataset) as f:\n",
    "    next(f) # skip the header row\n",
    "    next(f)\n",
    "    for line in f:\n",
    "        fields = line.split(',')\n",
    "        vector_vix.append(float(fields[6]))\n",
    "        \n",
    "\n",
    "# convert the vector to a 2D matrix\n",
    "matrix_vix = convertSeriesToMatrix(vector_vix, sequence_length)\n",
    "\n",
    "# shift all data by mean\n",
    "matrix_vix = np.array(matrix_vix)\n",
    "shifted_value = matrix_vix.mean()\n",
    "matrix_vix -= shifted_value\n",
    "print(\"Data shape: \", matrix_vix.shape)\n",
    "\n",
    "# split dataset: 80% for training and 10% for testing\n",
    "train_row = int(round(0.8 * matrix_vix.shape[0]))\n",
    "train_set = matrix_vix[:train_row, :]\n",
    "\n",
    "# shuffle the training set (but do not shuffle the test set)\n",
    "np.random.shuffle(train_set)\n",
    "\n",
    "# the training set\n",
    "X_train = train_set[:, :-1]\n",
    "# the last column is the true value to compute the mean-squared-error loss\n",
    "y_train = train_set[:, -1] \n",
    "# the test set\n",
    "X_test = matrix_vix[train_row:, :-1]\n",
    "y_test = matrix_vix[train_row:, -1]\n",
    "\n",
    "# the input to LSTM layer needs to have the shape of (number of samples, the dimension of each element)\n",
    "X_train = np.reshape(X_train, (X_train.shape[0], X_train.shape[1], 1))\n",
    "X_test = np.reshape(X_test, (X_test.shape[0], X_test.shape[1], 1))\n",
    "model = Sequential()\n",
    "# layer 1: LSTM\n",
    "model.add(LSTM( input_dim=1, output_dim=50, return_sequences=True))\n",
    "model.add(Dropout(0.2))\n",
    "# layer 2: LSTM\n",
    "model.add(LSTM(output_dim=100, return_sequences=False))\n",
    "model.add(Dropout(0.2))\n",
    "# layer 3: dense\n",
    "# linear activation: a(x) = x\n",
    "model.add(Dense(output_dim=1, activation='linear'))\n",
    "# compile the model\n",
    "model.compile(loss=\"mse\", optimizer=\"rmsprop\")\n",
    "model.fit(X_train, y_train, batch_size=512, epochs=100, validation_split=0.05, verbose=1)\n",
    "\n",
    "# evaluate the result\n",
    "test_mse = model.evaluate(X_test, y_test, verbose=1)\n",
    "print('The mean squared error (MSE) on the test data set is %.3f over %d test samples.' % (test_mse, len(y_test)))\n",
    "# get the predicted values\n",
    "predicted_values = model.predict(X_test)\n",
    "num_test_samples = len(predicted_values)\n",
    "predicted_values = np.reshape(predicted_values, (num_test_samples,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the results\n",
    "fig = plt.figure(figsize=(10,6))\n",
    "plt.plot(y_test + shifted_value)\n",
    "plt.plot(predicted_values + shifted_value)\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Price')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LSTM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> ### Model Artchitecture taken from Ralph Schlosser: https://www.kaggle.com/ternaryrealm/lstm-time-series-explorations-with-keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's load the required libs.\n",
    "# We'll be using the Tensorflow backend (default).\n",
    "from keras.models import Sequential\n",
    "from keras.layers.recurrent import LSTM\n",
    "from keras.layers.core import Dense, Activation, Dropout\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.utils import shuffle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.],\n",
       "       [0.],\n",
       "       [0.],\n",
       "       [0.],\n",
       "       [0.]], dtype=float32)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "data_raw = df['VolatilityForValuation'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n"
     ]
    }
   ],
   "source": [
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# FIXME: This helper function should be rewritten using numpy's shift function. See below.\n",
    "def create_dataset(dataset, window_size = 1):\n",
    "    data_X, data_Y = [], []\n",
    "    for i in range(len(dataset) - window_size - 1):\n",
    "        a = dataset[i:(i + window_size), 0]\n",
    "        data_X.append(a)\n",
    "        data_Y.append(dataset[i + window_size, 0])\n",
    "    return(np.array(data_X), np.array(data_Y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n"
     ]
    }
   ],
   "source": [
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fit_model(train_X, train_Y, window_size = 1):\n",
    "    model = Sequential()\n",
    "    \n",
    "    model.add(LSTM(4, \n",
    "                   input_shape = (1, window_size)))\n",
    "    model.add(Dense(1))\n",
    "    model.compile(loss = \"mean_squared_error\", \n",
    "                  optimizer = \"adam\")\n",
    "    model.fit(train_X, \n",
    "              train_Y, \n",
    "              epochs = 3, \n",
    "              batch_size = 1, \n",
    "              verbose = 2)\n",
    "    \n",
    "    return(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/3\n",
      " - 121s - loss: 4.6586e-04\n",
      "Epoch 2/3\n",
      " - 122s - loss: 2.9475e-04\n",
      "Epoch 3/3\n",
      " - 121s - loss: 2.9369e-04\n"
     ]
    }
   ],
   "source": [
    "# Fit the first model.\n",
    "model1 = fit_model(train_X, train_Y, window_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "\n",
    "def predict_and_score(model, X, Y):\n",
    "    # Make predictions on the original scale of the data.\n",
    "    pred = scaler.inverse_transform(model.predict(X))\n",
    "    # Prepare Y data to also be on the original scale for interpretability.\n",
    "    orig_data = scaler.inverse_transform([Y])\n",
    "    # Calculate RMSE.\n",
    "    score = math.sqrt(mean_squared_error(orig_data[0], pred[:, 0]))\n",
    "    return(score, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training data score: 4.52 RMSE\n",
      "Test data score: 5.65 RMSE\n"
     ]
    }
   ],
   "source": [
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict\n",
    "\n",
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"VolatilityForValuation\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.],\n",
       "       [0.],\n",
       "       [0.],\n",
       "       [0.],\n",
       "       [0.]], dtype=float32)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "df['Open'] = pd.to_numeric(df['Open'], errors='coerce')\n",
    "data_raw = df['Open'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n"
     ]
    }
   ],
   "source": [
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n"
     ]
    }
   ],
   "source": [
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/3\n",
      " - 124s - loss: 5.6262e-05\n",
      "Epoch 2/3\n",
      " - 124s - loss: 5.4933e-05\n",
      "Epoch 3/3\n",
      " - 122s - loss: 5.5274e-05\n"
     ]
    }
   ],
   "source": [
    "model1 = fit_model(train_X, train_Y, window_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training data score: 39.47 RMSE\n",
      "Test data score: 26.14 RMSE\n"
     ]
    }
   ],
   "source": [
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict\n",
    "\n",
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Open Price\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Epoch 1/3\n",
      " - 126s - loss: 5.4433e-05\n",
      "Epoch 2/3\n",
      " - 123s - loss: 5.5245e-05\n",
      "Epoch 3/3\n",
      " - 125s - loss: 5.4864e-05\n",
      "Training data score: 38.80 RMSE\n",
      "Test data score: 26.04 RMSE\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "df['Close'] = pd.to_numeric(df['Close'], errors='coerce')\n",
    "data_raw = df['Close'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]\n",
    "\n",
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "model1 = fit_model(train_X, train_Y, window_size)\n",
    "\n",
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)\n",
    "\n",
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict\n",
    "\n",
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Close Price\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Epoch 1/3\n",
      " - 127s - loss: 5.5524e-05\n",
      "Epoch 2/3\n",
      " - 124s - loss: 5.5179e-05\n",
      "Epoch 3/3\n",
      " - 125s - loss: 5.4904e-05\n",
      "Training data score: 39.50 RMSE\n",
      "Test data score: 26.31 RMSE\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "df['High'] = pd.to_numeric(df['High'], errors='coerce')\n",
    "data_raw = df['High'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]\n",
    "\n",
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "model1 = fit_model(train_X, train_Y, window_size)\n",
    "\n",
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)\n",
    "\n",
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict\n",
    "\n",
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"High Price\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Epoch 1/3\n",
      " - 126s - loss: 5.5179e-05\n",
      "Epoch 2/3\n",
      " - 126s - loss: 5.6223e-05\n",
      "Epoch 3/3\n",
      " - 127s - loss: 5.5461e-05\n",
      "Training data score: 38.79 RMSE\n",
      "Test data score: 26.03 RMSE\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "df['Low'] = pd.to_numeric(df['Low'], errors='coerce')\n",
    "data_raw = df['Low'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]\n",
    "\n",
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "model1 = fit_model(train_X, train_Y, window_size)\n",
    "\n",
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)\n",
    "\n",
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict\n",
    "\n",
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Low Price\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Epoch 1/3\n",
      " - 129s - loss: 1.4095e-04\n",
      "Epoch 2/3\n",
      " - 129s - loss: 1.1408e-04\n",
      "Epoch 3/3\n",
      " - 128s - loss: 1.1305e-04\n",
      "Training data score: 1088.28 RMSE\n",
      "Test data score: 3650.44 RMSE\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "df['ExercisePrice'] = pd.to_numeric(df['ExercisePrice'], errors='coerce')\n",
    "data_raw = df['ExercisePrice'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]\n",
    "\n",
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "model1 = fit_model(train_X, train_Y, window_size)\n",
    "\n",
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)\n",
    "\n",
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict\n",
    "\n",
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"ExercisePrice\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of entries (training set, test set): (60536, 40358)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Original training data shape:\n",
      "(60534, 1)\n",
      "New training data shape:\n",
      "(60534, 1, 1)\n",
      "Epoch 1/3\n",
      " - 131s - loss: 4.8680e-04\n",
      "Epoch 2/3\n",
      " - 130s - loss: 4.2321e-04\n",
      "Epoch 3/3\n",
      " - 129s - loss: 4.1138e-04\n",
      "Training data score: 150.90 RMSE\n",
      "Test data score: 71.92 RMSE\n"
     ]
    }
   ],
   "source": [
    "# Get the raw data values from the pandas data frame.\n",
    "df['Settlement'] = pd.to_numeric(df['Settlement'], errors='coerce')\n",
    "data_raw = df['Settlement'].fillna(0).values.astype(\"float32\")\n",
    "\n",
    "# We apply the MinMax scaler from sklearn\n",
    "# to normalize data in the (0, 1) interval.\n",
    "scaler = MinMaxScaler(feature_range = (0, 1))\n",
    "dataset = scaler.fit_transform(data_raw.reshape(-1, 1))\n",
    "\n",
    "# Print a few values.\n",
    "dataset[0:5]\n",
    "\n",
    "# Using 60% of data for training, 40% for validation.\n",
    "TRAIN_SIZE = 0.60\n",
    "\n",
    "train_size = int(len(dataset) * TRAIN_SIZE)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size, :], dataset[train_size:len(dataset), :]\n",
    "print(\"Number of entries (training set, test set): \" + str((len(train), len(test))))\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Create test and training sets for one-step-ahead regression.\n",
    "window_size = 1\n",
    "train_X, train_Y = create_dataset(train, window_size)\n",
    "test_X, test_Y = create_dataset(test, window_size)\n",
    "print(\"Original training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "# Reshape the input data into appropriate form for Keras.\n",
    "train_X = np.reshape(train_X, (train_X.shape[0], 1, train_X.shape[1]))\n",
    "test_X = np.reshape(test_X, (test_X.shape[0], 1, test_X.shape[1]))\n",
    "print(\"New training data shape:\")\n",
    "print(train_X.shape)\n",
    "\n",
    "model1 = fit_model(train_X, train_Y, window_size)\n",
    "\n",
    "rmse_train, train_predict = predict_and_score(model1, train_X, train_Y)\n",
    "rmse_test, test_predict = predict_and_score(model1, test_X, test_Y)\n",
    "\n",
    "print(\"Training data score: %.2f RMSE\" % rmse_train)\n",
    "print(\"Test data score: %.2f RMSE\" % rmse_test)\n",
    "\n",
    "# Start with training predictions.\n",
    "train_predict_plot = np.empty_like(dataset)\n",
    "train_predict_plot[:, :] = np.nan\n",
    "train_predict_plot[window_size:len(train_predict) + window_size, :] = train_predict\n",
    "\n",
    "# Add test predictions.\n",
    "test_predict_plot = np.empty_like(dataset)\n",
    "test_predict_plot[:, :] = np.nan\n",
    "test_predict_plot[len(train_predict) + (window_size * 2) + 1:len(dataset) - 1, :] = test_predict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create the plot.\n",
    "fig, ax = plt.subplots(figsize=(12,5))\n",
    "#ax.plot(x, y)\n",
    "#plt.figure(figsize = (15, 5))\n",
    "ax.plot(scaler.inverse_transform(dataset), label = \"True value\")\n",
    "ax.plot(train_predict_plot, label = \"Training set prediction\")\n",
    "ax.plot(test_predict_plot, label = \"Test set prediction\")\n",
    "plt.xlabel(\"Time\")\n",
    "plt.ylabel(\"Settlement\")\n",
    "plt.title(\"Comparison true vs. predicted training / test\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EuronextLisbon-IntradayTrades-20190520.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>Internal code;ISIN code;Instrument name;Trading Place;Category;Issuing country code;Segment;Trading currency code;Trading unit (currency or %);Registred date;Trade number;Trade indicator;Trade date;Trade time;Cross-trade indicator;Price (in trading currency);Volume;Turnover in euro</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>283259;PTBIXRYE0025;BCP IBEX 35       ;Lisbon;...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1364220;PTMENWOM0007;MOTA 4.5% 28NOV22 ;Lisbon...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>377693;PTBIWWYE0052;BCP OURO          ;Lisbon;...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1456889;DE000CJ36V67;DAX  11700TL 0619Z;Lisbon...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1281017;PTOTVMOE0000;OTRV FRN 23JUL25  ;Lisbon...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Internal code;ISIN code;Instrument name;Trading Place;Category;Issuing country code;Segment;Trading currency code;Trading unit (currency or %);Registred date;Trade number;Trade indicator;Trade date;Trade time;Cross-trade indicator;Price (in trading currency);Volume;Turnover in euro\n",
       "0  283259;PTBIXRYE0025;BCP IBEX 35       ;Lisbon;...                                                                                                                                                                                                                                        \n",
       "1  1364220;PTMENWOM0007;MOTA 4.5% 28NOV22 ;Lisbon...                                                                                                                                                                                                                                        \n",
       "2  377693;PTBIWWYE0052;BCP OURO          ;Lisbon;...                                                                                                                                                                                                                                        \n",
       "3  1456889;DE000CJ36V67;DAX  11700TL 0619Z;Lisbon...                                                                                                                                                                                                                                        \n",
       "4  1281017;PTOTVMOE0000;OTRV FRN 23JUL25  ;Lisbon...                                                                                                                                                                                                                                        "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = pd.read_csv(\"../input/level2stockanalysis/EuronextLisbon-IntradayTrades-20190520.csv\")\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Internal code', 'ISIN code', 'Instrument name', 'Trading Place', 'Category', 'Issuing country code', 'Segment', 'Trading currency code', 'Trading unit (currency or %)', 'Registred date', 'Trade number', 'Trade indicator', 'Trade date', 'Trade time', 'Cross-trade indicator', 'Price (in trading currency)', 'Volume', 'Turnover in euro']\n"
     ]
    }
   ],
   "source": [
    "new_header = \"Internal code;ISIN code;Instrument name;Trading Place;Category;Issuing country code;Segment;Trading currency code;Trading unit (currency or %);Registred date;Trade number;Trade indicator;Trade date;Trade time;Cross-trade indicator;Price (in trading currency);Volume;Turnover in euro\"\n",
    "new_header = new_header.split(';')\n",
    "print(new_header)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>Internal code</th>\n",
       "      <th>ISIN code</th>\n",
       "      <th>Instrument name</th>\n",
       "      <th>Trading Place</th>\n",
       "      <th>Category</th>\n",
       "      <th>Issuing country code</th>\n",
       "      <th>Segment</th>\n",
       "      <th>Trading currency code</th>\n",
       "      <th>Trading unit (currency or %)</th>\n",
       "      <th>Registred date</th>\n",
       "      <th>Trade number</th>\n",
       "      <th>Trade indicator</th>\n",
       "      <th>Trade date</th>\n",
       "      <th>Trade time</th>\n",
       "      <th>Cross-trade indicator</th>\n",
       "      <th>Price (in trading currency)</th>\n",
       "      <th>Volume</th>\n",
       "      <th>Turnover in euro</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>283259</td>\n",
       "      <td>PTBIXRYE0025</td>\n",
       "      <td>BCP IBEX 35</td>\n",
       "      <td>Lisbon</td>\n",
       "      <td>Certificate</td>\n",
       "      <td>PRT</td>\n",
       "      <td></td>\n",
       "      <td>EUR</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>272969</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>13:47:18:871798</td>\n",
       "      <td>0</td>\n",
       "      <td>92.05</td>\n",
       "      <td>30</td>\n",
       "      <td>2761.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1364220</td>\n",
       "      <td>PTMENWOM0007</td>\n",
       "      <td>MOTA 4.5% 28NOV22</td>\n",
       "      <td>Lisbon</td>\n",
       "      <td>Other bond</td>\n",
       "      <td>PRT</td>\n",
       "      <td></td>\n",
       "      <td>EUR</td>\n",
       "      <td>2</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>857852</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>14:30:00:004348</td>\n",
       "      <td>0</td>\n",
       "      <td>101.76</td>\n",
       "      <td>12500</td>\n",
       "      <td>12720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>377693</td>\n",
       "      <td>PTBIWWYE0052</td>\n",
       "      <td>BCP OURO</td>\n",
       "      <td>Lisbon</td>\n",
       "      <td>Certificate</td>\n",
       "      <td>PRT</td>\n",
       "      <td></td>\n",
       "      <td>EUR</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>14697</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>15:21:58:804471</td>\n",
       "      <td>0</td>\n",
       "      <td>106.1</td>\n",
       "      <td>10</td>\n",
       "      <td>1061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1456889</td>\n",
       "      <td>DE000CJ36V67</td>\n",
       "      <td>DAX  11700TL 0619Z</td>\n",
       "      <td>Lisbon</td>\n",
       "      <td>Warrant</td>\n",
       "      <td>DEU</td>\n",
       "      <td></td>\n",
       "      <td>EUR</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>223044</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>10:31:53:064227</td>\n",
       "      <td>0</td>\n",
       "      <td>2.08</td>\n",
       "      <td>1250</td>\n",
       "      <td>2600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1281017</td>\n",
       "      <td>PTOTVMOE0000</td>\n",
       "      <td>OTRV FRN 23JUL25</td>\n",
       "      <td>Lisbon</td>\n",
       "      <td>Government bond</td>\n",
       "      <td>PRT</td>\n",
       "      <td></td>\n",
       "      <td>EUR</td>\n",
       "      <td>2</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>858309</td>\n",
       "      <td>1</td>\n",
       "      <td>2019-05-20</td>\n",
       "      <td>12:25:16:744907</td>\n",
       "      <td>0</td>\n",
       "      <td>103.79</td>\n",
       "      <td>3000</td>\n",
       "      <td>3113.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Internal code     ISIN code       ...        Volume Turnover in euro\n",
       "0        283259  PTBIXRYE0025       ...            30           2761.5\n",
       "1       1364220  PTMENWOM0007       ...         12500            12720\n",
       "2        377693  PTBIWWYE0052       ...            10             1061\n",
       "3       1456889  DE000CJ36V67       ...          1250             2600\n",
       "4       1281017  PTOTVMOE0000       ...          3000           3113.7\n",
       "\n",
       "[5 rows x 18 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = df2.iloc[:,0].str.split(';', expand=True)\n",
    "\n",
    "df2.columns = new_header #set the header row as the df header\n",
    "\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16075, 18)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['Internal code', 'ISIN code', 'Instrument name', 'Trading Place',\n",
      "       'Category', 'Issuing country code', 'Segment', 'Trading currency code',\n",
      "       'Trading unit (currency or %)', 'Registred date', 'Trade number',\n",
      "       'Trade indicator', 'Trade date', 'Trade time', 'Cross-trade indicator',\n",
      "       'Price (in trading currency)', 'Volume', 'Turnover in euro'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(df2.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>percent_missing</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>D</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MarketPlace</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AssetClass</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AssetGroup</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExchangeCode</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InstrumentType</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InstrumentTrading Symbol</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>InstrumentID</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExpiryDate</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExercisePrice</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ActualExercisePrice</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OptionCategory</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Open</th>\n",
       "      <td>35.395564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>High</th>\n",
       "      <td>35.395564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Low</th>\n",
       "      <td>35.395564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TradeCount</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Close</th>\n",
       "      <td>35.395564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Settlement</th>\n",
       "      <td>35.823736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COBVolume</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OffCOBVolume</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TotalVolume</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OpenInterest</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OpenInterestDate</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EndofDayDelta</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Volatility</th>\n",
       "      <td>0.000991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VolatilityForValuation</th>\n",
       "      <td>36.132971</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          percent_missing\n",
       "0                                        \n",
       "D                                0.000000\n",
       "MarketPlace                      0.000000\n",
       "AssetClass                       0.000000\n",
       "AssetGroup                       0.000991\n",
       "ExchangeCode                     0.000991\n",
       "InstrumentType                   0.000991\n",
       "InstrumentTrading Symbol         0.000991\n",
       "InstrumentID                     0.000991\n",
       "ExpiryDate                       0.000991\n",
       "ExercisePrice                    0.000991\n",
       "ActualExercisePrice              0.000991\n",
       "OptionCategory                   0.000991\n",
       "Open                            35.395564\n",
       "High                            35.395564\n",
       "Low                             35.395564\n",
       "TradeCount                       0.000991\n",
       "Close                           35.395564\n",
       "Settlement                      35.823736\n",
       "COBVolume                        0.000991\n",
       "OffCOBVolume                     0.000991\n",
       "TotalVolume                      0.000991\n",
       "OpenInterest                     0.000991\n",
       "OpenInterestDate                 0.000991\n",
       "EndofDayDelta                    0.000991\n",
       "Volatility                       0.000991\n",
       "VolatilityForValuation          36.132971"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percent_missing = df.isnull().sum() * 100 / len(df)\n",
    "missing_values = pd.DataFrame({'percent_missing': percent_missing})\n",
    "missing_values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Visualisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df2['Turnover in euro'] = pd.to_numeric(df2['Turnover in euro'], errors='coerce')\n",
    "#df['ExercisePrice'].value_counts()\n",
    "df2['Turnover in euro'].plot()\n",
    "plt.title(\"Turnover in euro\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}