results.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predicting Stock Market Crashes -- Comparison of Results\n",
    "- Linear Regression / Logistic Regression / SVM (linear kernel) / SVM (RBF kernel) / Decision Trees / RNN LSTM\n",
    "- 16 features for all models except RNN / 2 sequences of length 14 for RNN LSTMs\n",
    "- Predictions for crash withinin 1 month / 3 months / 6 months"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from pylab import rcParams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 1 month prediction results:\n",
    "d = {}\n",
    "d['Linear Regression'] = [0.02, 0.05, 0.14, 0.4, 0.95, 0.29, 0.14,\\\n",
    "                          0.36, 0.95, 0.27, 0.02, 0.03, 0.16, 0.35, 0.96, 0.28]\n",
    "d['Logistic Regression'] = [0.02, 0.07, 0.11, 0.47, 0.93, 0.28, 0.11,\\\n",
    "                            0.42, 0.93, 0.26, 0.02, 0.07, 0.12, 0.45, 0.94, 0.29]\n",
    "d['SVC (linear)'] = [0.02, 0.06, 0.12, 0.48, 0.93, 0.30, 0.12, 0.44,\\\n",
    "                     0.93, 0.28, 0.02, 0.06, 0.12, 0.44, 0.94, 0.29]\n",
    "d['SVC (RBF)'] = [0.02, 0.08, 0.12, 0.60, 0.92, 0.33, 0.10, 0.49, 0.91,\\\n",
    "                  0.28, 0.02, 0.07, 0.09, 0.39, 0.93, 0.24]\n",
    "d['Decision Tree'] = [0.02, 0.01, 0.99, 0.90, 1.00, 0.92, 0.10, 0.10,\\\n",
    "                      0.97, 0.10, 0.02, 0.01, 0.11, 0.10, 0.97, 0.10]\n",
    "d['RNN LSTM'] = [0.02, 0.04, 0.18, 0.44, 0.96, 0.34, 0.17, 0.37, 0.95,\\\n",
    "                 0.29, 0.02, 0.04, 0.10, 0.28, 0.95, 0.21] \n",
    "d['random'] = [0.02, 0.05, 0.02, 0.05, 0.94, 0.04, 0.02, 0.05, 0.94, 0.04,\\\n",
    "              0.02, 0.05, 0.02, 0.05, 0.94, 0.04]\n",
    "results_1m = pd.DataFrame.from_dict(d, orient='index')\n",
    "results_1m.columns = ['positive_actual', 'positive_pred', 'precision_tr', 'recall_tr',\\\n",
    "                      'accuracy_tr', 'score_tr', 'precision_val','recall_val', 'accuracy_val',\\\n",
    "                      'score_val', 'positive_act', 'positive_pred', 'precision_t', 'recall_t',\\\n",
    "                      'accuracy_t', 'score_t']\n",
    "\n",
    "# 3 months prediction results:\n",
    "d = {}\n",
    "d['Linear Regression'] = [0.04, 0.17, 0.13, 0.5, 0.83, 0.31, 0.13, 0.5,\\\n",
    "                          0.83, 0.32, 0.04, 0.16, 0.15, 0.59, 0.84, 0.37]\n",
    "d['Logistic Regression'] = [0.04, 0.16, 0.13, 0.47, 0.84, 0.31, 0.13, 0.46,\\\n",
    "                            0.84, 0.30, 0.04, 0.15, 0.15, 0.56, 0.85, 0.36]\n",
    "d['SVC (linear)'] = [0.04, 0.14, 0.14, 0.48, 0.86, 0.33, 0.15, 0.46, 0.86,\\\n",
    "                     0.32, 0.04, 0.14, 0.16, 0.52, 0.87, 0.36]\n",
    "d['SVC (RBF)'] = [0.04, 0.14, 0.17, 0.57, 0.86, 0.39, 0.15, 0.45, 0.86, 0.31,\\\n",
    "                  0.04, 0.14, 0.16, 0.52, 0.87, 0.36]\n",
    "d['Decision Tree'] = [0.04, 0.04, 1.00, 0.99, 1.00, 0.99, 0.12, 0.12, 0.92,\\\n",
    "                      0.11, 0.04, 0.04, 0.07, 0.07, 0.92, 0.07]\n",
    "d['RNN LSTM'] = [0.04, 0.15, 0.14, 0.5, 0.84, 0.33, 0.15, 0.47, 0.84, 0.32,\\\n",
    "                 0.04, 0.13, 0.15, 0.48, 0.87, 0.33] \n",
    "d['random'] = [0.04, 0.18, 0.04, 0.18, 0.19, 0.11, 0.04, 0.18, 0.19, 0.11,\\\n",
    "              0.04, 0.18, 0.04, 0.18, 0.19, 0.11]\n",
    "results_3m = pd.DataFrame.from_dict(d, orient='index')\n",
    "results_3m.columns = ['positive_actual', 'positive_pred', 'precision_tr', 'recall_tr',\\\n",
    "                      'accuracy_tr', 'score_tr', 'precision_val','recall_val', 'accuracy_val',\\\n",
    "                      'score_val', 'positive_act', 'positive_pred', 'precision_t', 'recall_t',\\\n",
    "                      'accuracy_t', 'score_t']\n",
    "\n",
    "# 6 months prediction results:\n",
    "d = {}\n",
    "d['Linear Regression'] = [0.08, 0.37, 0.14, 0.61, 0.65, 0.36, 0.14, 0.63,\\\n",
    "                          0.66, 0.37, 0.07, 0.36, 0.14, 0.68, 0.67, 0.38]\n",
    "d['Logistic Regression'] = [0.08, 0.31, 0.15, 0.56, 0.7, 0.36, 0.15, 0.58,\\\n",
    "                            0.71, 0.36, 0.08, 0.31, 0.16, 0.66, 0.72, 0.40]\n",
    "d['SVC (linear)'] = [0.08, 0.28, 0.15, 0.55, 0.73, 0.36, 0.16, 0.55, 0.73,\\\n",
    "                     0.36, 0.08, 0.28, 0.17, 0.6, 0.77, 0.40]\n",
    "d['SVC (RBF)'] = [0.08, 0.32, 0.16, 0.67, 0.7, 0.42, 0.16, 0.62, 0.7, 0.38,\\\n",
    "                  0.08, 0.32, 0.15, 0.71, 0.69, 0.41]\n",
    "d['Decision Tree'] = [0.08, 0.08, 1.00, 1.00, 1.00, 1.00, 0.15, 0.13, 0.87,\\\n",
    "                      0.13, 0.08, 0.08, 0.12, 0.12, 0.88, 0.12]\n",
    "d['RNN LSTM'] = [0.08, 0.36, 0.15, 0.64, 0.66, 0.37, 0.14, 0.63, 0.66, 0.36,\\\n",
    "                 0.07, 0.35, 0.15, 0.71, 0.68, 0.40]\n",
    "d['random'] = [0.08, 0.36, 0.08, 0.36, 0.61, 0.21, 0.08, 0.36, 0.61, 0.21,\\\n",
    "              0.08, 0.36, 0.08, 0.36, 0.61, 0.21]\n",
    "results_6m = pd.DataFrame.from_dict(d, orient='index')\n",
    "results_6m.columns = ['positive_actual', 'positive_pred', 'precision_tr', 'recall_tr',\\\n",
    "                      'accuracy_tr', 'score_tr', 'precision_val','recall_val', 'accuracy_val',\\\n",
    "                      'score_val', 'positive_act', 'positive_pred', 'precision_t', 'recall_t',\\\n",
    "                      'accuracy_t', 'score_t']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training results 1 month prediciton:\n"
     ]
    },
    {
     "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>positive_actual</th>\n",
       "      <th>positive_pred</th>\n",
       "      <th>precision_tr</th>\n",
       "      <th>recall_tr</th>\n",
       "      <th>accuracy_tr</th>\n",
       "      <th>score_tr</th>\n",
       "      <th>precision_val</th>\n",
       "      <th>recall_val</th>\n",
       "      <th>accuracy_val</th>\n",
       "      <th>score_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Linear Regression</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.40</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.29</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Logistic Regression</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (linear)</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (RBF)</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.91</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.90</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNN LSTM</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     positive_actual  positive_pred  precision_tr  recall_tr  \\\n",
       "Linear Regression               0.02           0.05          0.14       0.40   \n",
       "Logistic Regression             0.02           0.07          0.11       0.47   \n",
       "SVC (linear)                    0.02           0.06          0.12       0.48   \n",
       "SVC (RBF)                       0.02           0.08          0.12       0.60   \n",
       "Decision Tree                   0.02           0.01          0.99       0.90   \n",
       "RNN LSTM                        0.02           0.04          0.18       0.44   \n",
       "random                          0.02           0.05          0.02       0.05   \n",
       "\n",
       "                     accuracy_tr  score_tr  precision_val  recall_val  \\\n",
       "Linear Regression           0.95      0.29           0.14        0.36   \n",
       "Logistic Regression         0.93      0.28           0.11        0.42   \n",
       "SVC (linear)                0.93      0.30           0.12        0.44   \n",
       "SVC (RBF)                   0.92      0.33           0.10        0.49   \n",
       "Decision Tree               1.00      0.92           0.10        0.10   \n",
       "RNN LSTM                    0.96      0.34           0.17        0.37   \n",
       "random                      0.94      0.04           0.02        0.05   \n",
       "\n",
       "                     accuracy_val  score_val  \n",
       "Linear Regression            0.95       0.27  \n",
       "Logistic Regression          0.93       0.26  \n",
       "SVC (linear)                 0.93       0.28  \n",
       "SVC (RBF)                    0.91       0.28  \n",
       "Decision Tree                0.97       0.10  \n",
       "RNN LSTM                     0.95       0.29  \n",
       "random                       0.94       0.04  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training results 3 month prediciton:\n"
     ]
    },
    {
     "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>positive_actual</th>\n",
       "      <th>positive_pred</th>\n",
       "      <th>precision_tr</th>\n",
       "      <th>recall_tr</th>\n",
       "      <th>accuracy_tr</th>\n",
       "      <th>score_tr</th>\n",
       "      <th>precision_val</th>\n",
       "      <th>recall_val</th>\n",
       "      <th>accuracy_val</th>\n",
       "      <th>score_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Linear Regression</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Logistic Regression</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.46</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (linear)</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.46</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (RBF)</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.39</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.45</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.04</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.99</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.99</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNN LSTM</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.33</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.47</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.19</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.19</td>\n",
       "      <td>0.11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     positive_actual  positive_pred  precision_tr  recall_tr  \\\n",
       "Linear Regression               0.04           0.17          0.13       0.50   \n",
       "Logistic Regression             0.04           0.16          0.13       0.47   \n",
       "SVC (linear)                    0.04           0.14          0.14       0.48   \n",
       "SVC (RBF)                       0.04           0.14          0.17       0.57   \n",
       "Decision Tree                   0.04           0.04          1.00       0.99   \n",
       "RNN LSTM                        0.04           0.15          0.14       0.50   \n",
       "random                          0.04           0.18          0.04       0.18   \n",
       "\n",
       "                     accuracy_tr  score_tr  precision_val  recall_val  \\\n",
       "Linear Regression           0.83      0.31           0.13        0.50   \n",
       "Logistic Regression         0.84      0.31           0.13        0.46   \n",
       "SVC (linear)                0.86      0.33           0.15        0.46   \n",
       "SVC (RBF)                   0.86      0.39           0.15        0.45   \n",
       "Decision Tree               1.00      0.99           0.12        0.12   \n",
       "RNN LSTM                    0.84      0.33           0.15        0.47   \n",
       "random                      0.19      0.11           0.04        0.18   \n",
       "\n",
       "                     accuracy_val  score_val  \n",
       "Linear Regression            0.83       0.32  \n",
       "Logistic Regression          0.84       0.30  \n",
       "SVC (linear)                 0.86       0.32  \n",
       "SVC (RBF)                    0.86       0.31  \n",
       "Decision Tree                0.92       0.11  \n",
       "RNN LSTM                     0.84       0.32  \n",
       "random                       0.19       0.11  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training results 6 month prediciton:\n"
     ]
    },
    {
     "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>positive_actual</th>\n",
       "      <th>positive_pred</th>\n",
       "      <th>precision_tr</th>\n",
       "      <th>recall_tr</th>\n",
       "      <th>accuracy_tr</th>\n",
       "      <th>score_tr</th>\n",
       "      <th>precision_val</th>\n",
       "      <th>recall_val</th>\n",
       "      <th>accuracy_val</th>\n",
       "      <th>score_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Linear Regression</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.65</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.63</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Logistic Regression</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (linear)</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.55</td>\n",
       "      <td>0.73</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.55</td>\n",
       "      <td>0.73</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (RBF)</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.32</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.70</td>\n",
       "      <td>0.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNN LSTM</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.64</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.37</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.63</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.21</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     positive_actual  positive_pred  precision_tr  recall_tr  \\\n",
       "Linear Regression               0.08           0.37          0.14       0.61   \n",
       "Logistic Regression             0.08           0.31          0.15       0.56   \n",
       "SVC (linear)                    0.08           0.28          0.15       0.55   \n",
       "SVC (RBF)                       0.08           0.32          0.16       0.67   \n",
       "Decision Tree                   0.08           0.08          1.00       1.00   \n",
       "RNN LSTM                        0.08           0.36          0.15       0.64   \n",
       "random                          0.08           0.36          0.08       0.36   \n",
       "\n",
       "                     accuracy_tr  score_tr  precision_val  recall_val  \\\n",
       "Linear Regression           0.65      0.36           0.14        0.63   \n",
       "Logistic Regression         0.70      0.36           0.15        0.58   \n",
       "SVC (linear)                0.73      0.36           0.16        0.55   \n",
       "SVC (RBF)                   0.70      0.42           0.16        0.62   \n",
       "Decision Tree               1.00      1.00           0.15        0.13   \n",
       "RNN LSTM                    0.66      0.37           0.14        0.63   \n",
       "random                      0.61      0.21           0.08        0.36   \n",
       "\n",
       "                     accuracy_val  score_val  \n",
       "Linear Regression            0.66       0.37  \n",
       "Logistic Regression          0.71       0.36  \n",
       "SVC (linear)                 0.73       0.36  \n",
       "SVC (RBF)                    0.70       0.38  \n",
       "Decision Tree                0.87       0.13  \n",
       "RNN LSTM                     0.66       0.36  \n",
       "random                       0.61       0.21  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test results 1 month prediciton:\n"
     ]
    },
    {
     "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>positive_act</th>\n",
       "      <th>positive_pred</th>\n",
       "      <th>precision_t</th>\n",
       "      <th>recall_t</th>\n",
       "      <th>accuracy_t</th>\n",
       "      <th>score_t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Linear Regression</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Logistic Regression</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.45</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (linear)</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (RBF)</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.09</td>\n",
       "      <td>0.39</td>\n",
       "      <td>0.93</td>\n",
       "      <td>0.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.97</td>\n",
       "      <td>0.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNN LSTM</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.95</td>\n",
       "      <td>0.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random</th>\n",
       "      <td>0.02</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     positive_act  positive_pred  precision_t  recall_t  \\\n",
       "Linear Regression            0.02           0.03         0.16      0.35   \n",
       "Logistic Regression          0.02           0.07         0.12      0.45   \n",
       "SVC (linear)                 0.02           0.06         0.12      0.44   \n",
       "SVC (RBF)                    0.02           0.07         0.09      0.39   \n",
       "Decision Tree                0.02           0.01         0.11      0.10   \n",
       "RNN LSTM                     0.02           0.04         0.10      0.28   \n",
       "random                       0.02           0.05         0.02      0.05   \n",
       "\n",
       "                     accuracy_t  score_t  \n",
       "Linear Regression          0.96     0.28  \n",
       "Logistic Regression        0.94     0.29  \n",
       "SVC (linear)               0.94     0.29  \n",
       "SVC (RBF)                  0.93     0.24  \n",
       "Decision Tree              0.97     0.10  \n",
       "RNN LSTM                   0.95     0.21  \n",
       "random                     0.94     0.04  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test results 3 month prediciton:\n"
     ]
    },
    {
     "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>positive_act</th>\n",
       "      <th>positive_pred</th>\n",
       "      <th>precision_t</th>\n",
       "      <th>recall_t</th>\n",
       "      <th>accuracy_t</th>\n",
       "      <th>score_t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Linear Regression</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Logistic Regression</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.85</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (linear)</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (RBF)</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.07</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNN LSTM</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.87</td>\n",
       "      <td>0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random</th>\n",
       "      <td>0.04</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.19</td>\n",
       "      <td>0.11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     positive_act  positive_pred  precision_t  recall_t  \\\n",
       "Linear Regression            0.04           0.16         0.15      0.59   \n",
       "Logistic Regression          0.04           0.15         0.15      0.56   \n",
       "SVC (linear)                 0.04           0.14         0.16      0.52   \n",
       "SVC (RBF)                    0.04           0.14         0.16      0.52   \n",
       "Decision Tree                0.04           0.04         0.07      0.07   \n",
       "RNN LSTM                     0.04           0.13         0.15      0.48   \n",
       "random                       0.04           0.18         0.04      0.18   \n",
       "\n",
       "                     accuracy_t  score_t  \n",
       "Linear Regression          0.84     0.37  \n",
       "Logistic Regression        0.85     0.36  \n",
       "SVC (linear)               0.87     0.36  \n",
       "SVC (RBF)                  0.87     0.36  \n",
       "Decision Tree              0.92     0.07  \n",
       "RNN LSTM                   0.87     0.33  \n",
       "random                     0.19     0.11  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test results 6 month prediciton:\n"
     ]
    },
    {
     "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>positive_act</th>\n",
       "      <th>positive_pred</th>\n",
       "      <th>precision_t</th>\n",
       "      <th>recall_t</th>\n",
       "      <th>accuracy_t</th>\n",
       "      <th>score_t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Linear Regression</th>\n",
       "      <td>0.07</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Logistic Regression</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.31</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.72</td>\n",
       "      <td>0.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (linear)</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.77</td>\n",
       "      <td>0.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC (RBF)</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.32</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.69</td>\n",
       "      <td>0.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.12</td>\n",
       "      <td>0.88</td>\n",
       "      <td>0.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RNN LSTM</th>\n",
       "      <td>0.07</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.71</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>random</th>\n",
       "      <td>0.08</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.08</td>\n",
       "      <td>0.36</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     positive_act  positive_pred  precision_t  recall_t  \\\n",
       "Linear Regression            0.07           0.36         0.14      0.68   \n",
       "Logistic Regression          0.08           0.31         0.16      0.66   \n",
       "SVC (linear)                 0.08           0.28         0.17      0.60   \n",
       "SVC (RBF)                    0.08           0.32         0.15      0.71   \n",
       "Decision Tree                0.08           0.08         0.12      0.12   \n",
       "RNN LSTM                     0.07           0.35         0.15      0.71   \n",
       "random                       0.08           0.36         0.08      0.36   \n",
       "\n",
       "                     accuracy_t  score_t  \n",
       "Linear Regression          0.67     0.38  \n",
       "Logistic Regression        0.72     0.40  \n",
       "SVC (linear)               0.77     0.40  \n",
       "SVC (RBF)                  0.69     0.41  \n",
       "Decision Tree              0.88     0.12  \n",
       "RNN LSTM                   0.68     0.40  \n",
       "random                     0.61     0.21  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Training results 1 month prediciton:')\n",
    "display(results_1m.iloc[:,:10])\n",
    "print('Training results 3 month prediciton:')\n",
    "display(results_3m.iloc[:,:10])\n",
    "print('Training results 6 month prediciton:')\n",
    "display(results_6m.iloc[:,:10])\n",
    "print('Test results 1 month prediciton:')\n",
    "display(results_1m.iloc[:,10:])\n",
    "print('Test results 3 month prediciton:')\n",
    "display(results_3m.iloc[:,10:])\n",
    "print('Test results 6 month prediciton:')\n",
    "display(results_6m.iloc[:,10:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ---------- Plot results recall vs precision ---------- #\n",
    "colors = ['royalblue', 'blueviolet', 'gold', 'coral', 'green', 'red', 'gray']\n",
    "markers = ['s', 'o', 'D']\n",
    "f = lambda m,c: plt.plot([],[],marker=m, color=c, ls=\"none\")[0]\n",
    "handles = [f('s', colors[i]) for i in range(7)]\n",
    "handles += [f(markers[i], 'k') for i in range(3)]\n",
    "labels = list(results_6m.index) + ['1 month prediction', '3 month prediction', '6 month prediction']\n",
    "rcParams.update({'font.size': 16})\n",
    "rcParams['figure.figsize'] = 10, 8\n",
    "plt.style.use('seaborn-darkgrid')\n",
    "x = list(results_1m['recall_t'])\n",
    "y = list(results_1m['precision_t'])\n",
    "for x_, y_, c_ in zip(x, y, colors):\n",
    "    plt.scatter(x_, y_, marker='s', c=c_, linewidths=1, s=110, alpha=0.7)\n",
    "x = list(results_3m['recall_t'])\n",
    "y = list(results_3m['precision_t'])\n",
    "for x_, y_, c_ in zip(x, y, colors):\n",
    "    plt.scatter(x_, y_, marker='o', c=c_, linewidths=1, s=120, alpha=0.7)\n",
    "x = list(results_6m['recall_t'])\n",
    "y = list(results_6m['precision_t'])\n",
    "for x_, y_, c_ in zip(x, y, colors):\n",
    "    plt.scatter(x_, y_, marker='D', c=c_, linewidths=1, s=100, alpha=0.7)\n",
    "plt.legend(handles, labels, loc=4, framealpha=1)\n",
    "plt.title('Recall vs Precision all models')\n",
    "plt.xlim(xmin=0, xmax=1)\n",
    "plt.ylim(ymin=0, ymax=0.2)\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ---------- Plot results f-beta score ---------- #\n",
    "rcParams['figure.figsize'] = 10, 8\n",
    "rcParams.update({'font.size': 18})\n",
    "df = pd.concat([results_6m['score_t'], results_3m['score_t'], results_1m['score_t']], axis=1)\n",
    "df.columns = ['6 month prediciton', '3 month prediciton', '1 month prediciton']\n",
    "df = df.iloc[::-1]\n",
    "df.plot(kind='barh', stacked=False, width=0.75, color=['mediumaquamarine','mediumseagreen','green'])\n",
    "plt.title('F-beta score all models')\n",
    "plt.xlabel('F-beta score (beta = 2)')\n",
    "plt.legend(loc='best', bbox_to_anchor=(1, 0.5))\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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}