Domain_Adapted_Feature_Selection_Tech (1).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cdf0afcf",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "de52b4e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import silhouette_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "80c825e3",
   "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>Year</th>\n",
       "      <th>RPI</th>\n",
       "      <th>INDPRO</th>\n",
       "      <th>CE16OV</th>\n",
       "      <th>UNRATE</th>\n",
       "      <th>PAYEMS</th>\n",
       "      <th>USGOOD</th>\n",
       "      <th>USTPU</th>\n",
       "      <th>HOUST</th>\n",
       "      <th>PERMIT</th>\n",
       "      <th>...</th>\n",
       "      <th>EXUSUK</th>\n",
       "      <th>WPSFD49207</th>\n",
       "      <th>WPSID61</th>\n",
       "      <th>CPIAUCSL</th>\n",
       "      <th>SP500</th>\n",
       "      <th>NASDAQ</th>\n",
       "      <th>GOLDBAR</th>\n",
       "      <th>Regime</th>\n",
       "      <th>P/E</th>\n",
       "      <th>Dividend Yield</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1996-01</td>\n",
       "      <td>8909.327</td>\n",
       "      <td>74.6841</td>\n",
       "      <td>125125</td>\n",
       "      <td>5.6</td>\n",
       "      <td>118316</td>\n",
       "      <td>23196</td>\n",
       "      <td>23947</td>\n",
       "      <td>1467</td>\n",
       "      <td>1387</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5288</td>\n",
       "      <td>129.7</td>\n",
       "      <td>125.5</td>\n",
       "      <td>154.700</td>\n",
       "      <td>636.020020</td>\n",
       "      <td>1059.790039</td>\n",
       "      <td>20.754421</td>\n",
       "      <td>Normal</td>\n",
       "      <td>18.6845</td>\n",
       "      <td>2.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1996-02</td>\n",
       "      <td>8983.863</td>\n",
       "      <td>75.8344</td>\n",
       "      <td>125639</td>\n",
       "      <td>5.5</td>\n",
       "      <td>118739</td>\n",
       "      <td>23280</td>\n",
       "      <td>23988</td>\n",
       "      <td>1491</td>\n",
       "      <td>1420</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5360</td>\n",
       "      <td>129.7</td>\n",
       "      <td>125.0</td>\n",
       "      <td>155.000</td>\n",
       "      <td>640.429993</td>\n",
       "      <td>1100.050049</td>\n",
       "      <td>21.372637</td>\n",
       "      <td>Normal</td>\n",
       "      <td>18.8140</td>\n",
       "      <td>2.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1996-03</td>\n",
       "      <td>9015.588</td>\n",
       "      <td>75.7631</td>\n",
       "      <td>125862</td>\n",
       "      <td>5.5</td>\n",
       "      <td>118993</td>\n",
       "      <td>23276</td>\n",
       "      <td>24030</td>\n",
       "      <td>1424</td>\n",
       "      <td>1437</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5271</td>\n",
       "      <td>130.5</td>\n",
       "      <td>125.3</td>\n",
       "      <td>155.500</td>\n",
       "      <td>645.500000</td>\n",
       "      <td>1101.400024</td>\n",
       "      <td>21.460953</td>\n",
       "      <td>Normal</td>\n",
       "      <td>18.9630</td>\n",
       "      <td>2.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1996-04</td>\n",
       "      <td>9039.466</td>\n",
       "      <td>76.4562</td>\n",
       "      <td>125994</td>\n",
       "      <td>5.6</td>\n",
       "      <td>119158</td>\n",
       "      <td>23316</td>\n",
       "      <td>24043</td>\n",
       "      <td>1516</td>\n",
       "      <td>1463</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5160</td>\n",
       "      <td>130.9</td>\n",
       "      <td>125.7</td>\n",
       "      <td>156.100</td>\n",
       "      <td>654.169983</td>\n",
       "      <td>1190.520020</td>\n",
       "      <td>21.637589</td>\n",
       "      <td>Normal</td>\n",
       "      <td>18.7388</td>\n",
       "      <td>2.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1996-05</td>\n",
       "      <td>9078.928</td>\n",
       "      <td>77.0161</td>\n",
       "      <td>126244</td>\n",
       "      <td>5.6</td>\n",
       "      <td>119486</td>\n",
       "      <td>23358</td>\n",
       "      <td>24137</td>\n",
       "      <td>1504</td>\n",
       "      <td>1457</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5152</td>\n",
       "      <td>130.9</td>\n",
       "      <td>126.2</td>\n",
       "      <td>156.400</td>\n",
       "      <td>669.119995</td>\n",
       "      <td>1243.430054</td>\n",
       "      <td>22.255806</td>\n",
       "      <td>Normal</td>\n",
       "      <td>19.1670</td>\n",
       "      <td>2.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>2020-02</td>\n",
       "      <td>17219.643</td>\n",
       "      <td>109.2966</td>\n",
       "      <td>158759</td>\n",
       "      <td>3.5</td>\n",
       "      <td>152463</td>\n",
       "      <td>21205</td>\n",
       "      <td>27830</td>\n",
       "      <td>1567</td>\n",
       "      <td>1438</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2953</td>\n",
       "      <td>206.0</td>\n",
       "      <td>195.2</td>\n",
       "      <td>259.050</td>\n",
       "      <td>2954.219971</td>\n",
       "      <td>8567.370117</td>\n",
       "      <td>18.868309</td>\n",
       "      <td>Normal</td>\n",
       "      <td>25.3952</td>\n",
       "      <td>1.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>2020-03</td>\n",
       "      <td>16947.234</td>\n",
       "      <td>104.5221</td>\n",
       "      <td>155772</td>\n",
       "      <td>4.4</td>\n",
       "      <td>151090</td>\n",
       "      <td>21086</td>\n",
       "      <td>27723</td>\n",
       "      <td>1269</td>\n",
       "      <td>1356</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2369</td>\n",
       "      <td>202.4</td>\n",
       "      <td>192.2</td>\n",
       "      <td>257.953</td>\n",
       "      <td>2584.590088</td>\n",
       "      <td>7700.100098</td>\n",
       "      <td>18.215258</td>\n",
       "      <td>Recession</td>\n",
       "      <td>22.2177</td>\n",
       "      <td>2.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>2020-04</td>\n",
       "      <td>19124.430</td>\n",
       "      <td>91.2658</td>\n",
       "      <td>133403</td>\n",
       "      <td>14.7</td>\n",
       "      <td>130303</td>\n",
       "      <td>18698</td>\n",
       "      <td>24475</td>\n",
       "      <td>934</td>\n",
       "      <td>1066</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2420</td>\n",
       "      <td>196.3</td>\n",
       "      <td>185.3</td>\n",
       "      <td>255.902</td>\n",
       "      <td>2912.429932</td>\n",
       "      <td>8889.549805</td>\n",
       "      <td>25.572948</td>\n",
       "      <td>Recession</td>\n",
       "      <td>25.0359</td>\n",
       "      <td>2.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>2020-05</td>\n",
       "      <td>18296.406</td>\n",
       "      <td>92.0613</td>\n",
       "      <td>137242</td>\n",
       "      <td>13.3</td>\n",
       "      <td>133028</td>\n",
       "      <td>19374</td>\n",
       "      <td>24858</td>\n",
       "      <td>1038</td>\n",
       "      <td>1216</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2302</td>\n",
       "      <td>200.4</td>\n",
       "      <td>185.3</td>\n",
       "      <td>255.768</td>\n",
       "      <td>3044.310059</td>\n",
       "      <td>9489.870117</td>\n",
       "      <td>23.862782</td>\n",
       "      <td>Recession</td>\n",
       "      <td>26.1696</td>\n",
       "      <td>2.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>2020-06</td>\n",
       "      <td>17990.747</td>\n",
       "      <td>97.6032</td>\n",
       "      <td>142182</td>\n",
       "      <td>11.1</td>\n",
       "      <td>137809</td>\n",
       "      <td>19859</td>\n",
       "      <td>25852</td>\n",
       "      <td>1265</td>\n",
       "      <td>1258</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2523</td>\n",
       "      <td>201.1</td>\n",
       "      <td>187.7</td>\n",
       "      <td>257.214</td>\n",
       "      <td>3100.290039</td>\n",
       "      <td>10058.769530</td>\n",
       "      <td>26.864592</td>\n",
       "      <td>Recession</td>\n",
       "      <td>26.6508</td>\n",
       "      <td>1.95</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>293 rows × 45 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
       "0    1996-01   8909.327   74.6841  125125     5.6  118316   23196  23947   \n",
       "1    1996-02   8983.863   75.8344  125639     5.5  118739   23280  23988   \n",
       "2    1996-03   9015.588   75.7631  125862     5.5  118993   23276  24030   \n",
       "3    1996-04   9039.466   76.4562  125994     5.6  119158   23316  24043   \n",
       "4    1996-05   9078.928   77.0161  126244     5.6  119486   23358  24137   \n",
       "..       ...        ...       ...     ...     ...     ...     ...    ...   \n",
       "288  2020-02  17219.643  109.2966  158759     3.5  152463   21205  27830   \n",
       "289  2020-03  16947.234  104.5221  155772     4.4  151090   21086  27723   \n",
       "290  2020-04  19124.430   91.2658  133403    14.7  130303   18698  24475   \n",
       "291  2020-05  18296.406   92.0613  137242    13.3  133028   19374  24858   \n",
       "292  2020-06  17990.747   97.6032  142182    11.1  137809   19859  25852   \n",
       "\n",
       "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
       "0     1467    1387  ...  1.5288       129.7    125.5   154.700   636.020020   \n",
       "1     1491    1420  ...  1.5360       129.7    125.0   155.000   640.429993   \n",
       "2     1424    1437  ...  1.5271       130.5    125.3   155.500   645.500000   \n",
       "3     1516    1463  ...  1.5160       130.9    125.7   156.100   654.169983   \n",
       "4     1504    1457  ...  1.5152       130.9    126.2   156.400   669.119995   \n",
       "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
       "288   1567    1438  ...  1.2953       206.0    195.2   259.050  2954.219971   \n",
       "289   1269    1356  ...  1.2369       202.4    192.2   257.953  2584.590088   \n",
       "290    934    1066  ...  1.2420       196.3    185.3   255.902  2912.429932   \n",
       "291   1038    1216  ...  1.2302       200.4    185.3   255.768  3044.310059   \n",
       "292   1265    1258  ...  1.2523       201.1    187.7   257.214  3100.290039   \n",
       "\n",
       "           NASDAQ    GOLDBAR     Regime      P/E  Dividend Yield  \n",
       "0     1059.790039  20.754421     Normal  18.6845            2.26  \n",
       "1     1100.050049  21.372637     Normal  18.8140            2.16  \n",
       "2     1101.400024  21.460953     Normal  18.9630            2.18  \n",
       "3     1190.520020  21.637589     Normal  18.7388            2.19  \n",
       "4     1243.430054  22.255806     Normal  19.1670            2.15  \n",
       "..            ...        ...        ...      ...             ...  \n",
       "288   8567.370117  18.868309     Normal  25.3952            1.97  \n",
       "289   7700.100098  18.215258  Recession  22.2177            2.25  \n",
       "290   8889.549805  25.572948  Recession  25.0359            2.41  \n",
       "291   9489.870117  23.862782  Recession  26.1696            2.10  \n",
       "292  10058.769530  26.864592  Recession  26.6508            1.95  \n",
       "\n",
       "[293 rows x 45 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"D:\\\\Data\\Res.csv\")\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "4040bda8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier  \n",
    "\n",
    "data['Regime'] = data['Regime'].replace({'Normal': 0, 'Recession': 1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "88f990a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    263\n",
      "1     30\n",
      "Name: Regime, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "counts = data['Regime'].value_counts()\n",
    "print(counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "bdb8d7b5",
   "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>RPI</th>\n",
       "      <th>INDPRO</th>\n",
       "      <th>CE16OV</th>\n",
       "      <th>UNRATE</th>\n",
       "      <th>PAYEMS</th>\n",
       "      <th>USGOOD</th>\n",
       "      <th>USTPU</th>\n",
       "      <th>HOUST</th>\n",
       "      <th>PERMIT</th>\n",
       "      <th>DPCERA3M086SBEA</th>\n",
       "      <th>...</th>\n",
       "      <th>EXUSUK</th>\n",
       "      <th>WPSFD49207</th>\n",
       "      <th>WPSID61</th>\n",
       "      <th>CPIAUCSL</th>\n",
       "      <th>SP500</th>\n",
       "      <th>NASDAQ</th>\n",
       "      <th>GOLDBAR</th>\n",
       "      <th>Regime</th>\n",
       "      <th>P/E</th>\n",
       "      <th>Dividend Yield</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8909.327</td>\n",
       "      <td>74.6841</td>\n",
       "      <td>125125</td>\n",
       "      <td>5.6</td>\n",
       "      <td>118316</td>\n",
       "      <td>23196</td>\n",
       "      <td>23947</td>\n",
       "      <td>1467</td>\n",
       "      <td>1387</td>\n",
       "      <td>63.691</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5288</td>\n",
       "      <td>129.7</td>\n",
       "      <td>125.5</td>\n",
       "      <td>154.700</td>\n",
       "      <td>636.020020</td>\n",
       "      <td>1059.790039</td>\n",
       "      <td>20.754421</td>\n",
       "      <td>0</td>\n",
       "      <td>18.6845</td>\n",
       "      <td>2.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8983.863</td>\n",
       "      <td>75.8344</td>\n",
       "      <td>125639</td>\n",
       "      <td>5.5</td>\n",
       "      <td>118739</td>\n",
       "      <td>23280</td>\n",
       "      <td>23988</td>\n",
       "      <td>1491</td>\n",
       "      <td>1420</td>\n",
       "      <td>64.180</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5360</td>\n",
       "      <td>129.7</td>\n",
       "      <td>125.0</td>\n",
       "      <td>155.000</td>\n",
       "      <td>640.429993</td>\n",
       "      <td>1100.050049</td>\n",
       "      <td>21.372637</td>\n",
       "      <td>0</td>\n",
       "      <td>18.8140</td>\n",
       "      <td>2.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9015.588</td>\n",
       "      <td>75.7631</td>\n",
       "      <td>125862</td>\n",
       "      <td>5.5</td>\n",
       "      <td>118993</td>\n",
       "      <td>23276</td>\n",
       "      <td>24030</td>\n",
       "      <td>1424</td>\n",
       "      <td>1437</td>\n",
       "      <td>64.491</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5271</td>\n",
       "      <td>130.5</td>\n",
       "      <td>125.3</td>\n",
       "      <td>155.500</td>\n",
       "      <td>645.500000</td>\n",
       "      <td>1101.400024</td>\n",
       "      <td>21.460953</td>\n",
       "      <td>0</td>\n",
       "      <td>18.9630</td>\n",
       "      <td>2.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9039.466</td>\n",
       "      <td>76.4562</td>\n",
       "      <td>125994</td>\n",
       "      <td>5.6</td>\n",
       "      <td>119158</td>\n",
       "      <td>23316</td>\n",
       "      <td>24043</td>\n",
       "      <td>1516</td>\n",
       "      <td>1463</td>\n",
       "      <td>64.729</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5160</td>\n",
       "      <td>130.9</td>\n",
       "      <td>125.7</td>\n",
       "      <td>156.100</td>\n",
       "      <td>654.169983</td>\n",
       "      <td>1190.520020</td>\n",
       "      <td>21.637589</td>\n",
       "      <td>0</td>\n",
       "      <td>18.7388</td>\n",
       "      <td>2.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9078.928</td>\n",
       "      <td>77.0161</td>\n",
       "      <td>126244</td>\n",
       "      <td>5.6</td>\n",
       "      <td>119486</td>\n",
       "      <td>23358</td>\n",
       "      <td>24137</td>\n",
       "      <td>1504</td>\n",
       "      <td>1457</td>\n",
       "      <td>64.813</td>\n",
       "      <td>...</td>\n",
       "      <td>1.5152</td>\n",
       "      <td>130.9</td>\n",
       "      <td>126.2</td>\n",
       "      <td>156.400</td>\n",
       "      <td>669.119995</td>\n",
       "      <td>1243.430054</td>\n",
       "      <td>22.255806</td>\n",
       "      <td>0</td>\n",
       "      <td>19.1670</td>\n",
       "      <td>2.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>17219.643</td>\n",
       "      <td>109.2966</td>\n",
       "      <td>158759</td>\n",
       "      <td>3.5</td>\n",
       "      <td>152463</td>\n",
       "      <td>21205</td>\n",
       "      <td>27830</td>\n",
       "      <td>1567</td>\n",
       "      <td>1438</td>\n",
       "      <td>121.765</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2953</td>\n",
       "      <td>206.0</td>\n",
       "      <td>195.2</td>\n",
       "      <td>259.050</td>\n",
       "      <td>2954.219971</td>\n",
       "      <td>8567.370117</td>\n",
       "      <td>18.868309</td>\n",
       "      <td>0</td>\n",
       "      <td>25.3952</td>\n",
       "      <td>1.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>16947.234</td>\n",
       "      <td>104.5221</td>\n",
       "      <td>155772</td>\n",
       "      <td>4.4</td>\n",
       "      <td>151090</td>\n",
       "      <td>21086</td>\n",
       "      <td>27723</td>\n",
       "      <td>1269</td>\n",
       "      <td>1356</td>\n",
       "      <td>113.894</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2369</td>\n",
       "      <td>202.4</td>\n",
       "      <td>192.2</td>\n",
       "      <td>257.953</td>\n",
       "      <td>2584.590088</td>\n",
       "      <td>7700.100098</td>\n",
       "      <td>18.215258</td>\n",
       "      <td>1</td>\n",
       "      <td>22.2177</td>\n",
       "      <td>2.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>19124.430</td>\n",
       "      <td>91.2658</td>\n",
       "      <td>133403</td>\n",
       "      <td>14.7</td>\n",
       "      <td>130303</td>\n",
       "      <td>18698</td>\n",
       "      <td>24475</td>\n",
       "      <td>934</td>\n",
       "      <td>1066</td>\n",
       "      <td>99.932</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2420</td>\n",
       "      <td>196.3</td>\n",
       "      <td>185.3</td>\n",
       "      <td>255.902</td>\n",
       "      <td>2912.429932</td>\n",
       "      <td>8889.549805</td>\n",
       "      <td>25.572948</td>\n",
       "      <td>1</td>\n",
       "      <td>25.0359</td>\n",
       "      <td>2.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>18296.406</td>\n",
       "      <td>92.0613</td>\n",
       "      <td>137242</td>\n",
       "      <td>13.3</td>\n",
       "      <td>133028</td>\n",
       "      <td>19374</td>\n",
       "      <td>24858</td>\n",
       "      <td>1038</td>\n",
       "      <td>1216</td>\n",
       "      <td>108.449</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2302</td>\n",
       "      <td>200.4</td>\n",
       "      <td>185.3</td>\n",
       "      <td>255.768</td>\n",
       "      <td>3044.310059</td>\n",
       "      <td>9489.870117</td>\n",
       "      <td>23.862782</td>\n",
       "      <td>1</td>\n",
       "      <td>26.1696</td>\n",
       "      <td>2.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>17990.747</td>\n",
       "      <td>97.6032</td>\n",
       "      <td>142182</td>\n",
       "      <td>11.1</td>\n",
       "      <td>137809</td>\n",
       "      <td>19859</td>\n",
       "      <td>25852</td>\n",
       "      <td>1265</td>\n",
       "      <td>1258</td>\n",
       "      <td>114.880</td>\n",
       "      <td>...</td>\n",
       "      <td>1.2523</td>\n",
       "      <td>201.1</td>\n",
       "      <td>187.7</td>\n",
       "      <td>257.214</td>\n",
       "      <td>3100.290039</td>\n",
       "      <td>10058.769530</td>\n",
       "      <td>26.864592</td>\n",
       "      <td>1</td>\n",
       "      <td>26.6508</td>\n",
       "      <td>1.95</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>293 rows × 44 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  HOUST  \\\n",
       "0     8909.327   74.6841  125125     5.6  118316   23196  23947   1467   \n",
       "1     8983.863   75.8344  125639     5.5  118739   23280  23988   1491   \n",
       "2     9015.588   75.7631  125862     5.5  118993   23276  24030   1424   \n",
       "3     9039.466   76.4562  125994     5.6  119158   23316  24043   1516   \n",
       "4     9078.928   77.0161  126244     5.6  119486   23358  24137   1504   \n",
       "..         ...       ...     ...     ...     ...     ...    ...    ...   \n",
       "288  17219.643  109.2966  158759     3.5  152463   21205  27830   1567   \n",
       "289  16947.234  104.5221  155772     4.4  151090   21086  27723   1269   \n",
       "290  19124.430   91.2658  133403    14.7  130303   18698  24475    934   \n",
       "291  18296.406   92.0613  137242    13.3  133028   19374  24858   1038   \n",
       "292  17990.747   97.6032  142182    11.1  137809   19859  25852   1265   \n",
       "\n",
       "     PERMIT  DPCERA3M086SBEA  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL  \\\n",
       "0      1387           63.691  ...  1.5288       129.7    125.5   154.700   \n",
       "1      1420           64.180  ...  1.5360       129.7    125.0   155.000   \n",
       "2      1437           64.491  ...  1.5271       130.5    125.3   155.500   \n",
       "3      1463           64.729  ...  1.5160       130.9    125.7   156.100   \n",
       "4      1457           64.813  ...  1.5152       130.9    126.2   156.400   \n",
       "..      ...              ...  ...     ...         ...      ...       ...   \n",
       "288    1438          121.765  ...  1.2953       206.0    195.2   259.050   \n",
       "289    1356          113.894  ...  1.2369       202.4    192.2   257.953   \n",
       "290    1066           99.932  ...  1.2420       196.3    185.3   255.902   \n",
       "291    1216          108.449  ...  1.2302       200.4    185.3   255.768   \n",
       "292    1258          114.880  ...  1.2523       201.1    187.7   257.214   \n",
       "\n",
       "           SP500        NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
       "0     636.020020   1059.790039  20.754421       0  18.6845            2.26  \n",
       "1     640.429993   1100.050049  21.372637       0  18.8140            2.16  \n",
       "2     645.500000   1101.400024  21.460953       0  18.9630            2.18  \n",
       "3     654.169983   1190.520020  21.637589       0  18.7388            2.19  \n",
       "4     669.119995   1243.430054  22.255806       0  19.1670            2.15  \n",
       "..           ...           ...        ...     ...      ...             ...  \n",
       "288  2954.219971   8567.370117  18.868309       0  25.3952            1.97  \n",
       "289  2584.590088   7700.100098  18.215258       1  22.2177            2.25  \n",
       "290  2912.429932   8889.549805  25.572948       1  25.0359            2.41  \n",
       "291  3044.310059   9489.870117  23.862782       1  26.1696            2.10  \n",
       "292  3100.290039  10058.769530  26.864592       1  26.6508            1.95  \n",
       "\n",
       "[293 rows x 44 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.drop(\"Year\",axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8f70bf3c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "56c02537",
   "metadata": {},
   "outputs": [],
   "source": [
    "# DOMAIN 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "e38a1355",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['DTCTHFNM', 'T1YFFM', 'AAAFFM', 'TB3SMFFM']  \n",
    "target = 'Regime'\n",
    "\n",
    "\n",
    "data_1 = data[features + [target]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "39855971",
   "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>DTCTHFNM</th>\n",
       "      <th>T1YFFM</th>\n",
       "      <th>AAAFFM</th>\n",
       "      <th>TB3SMFFM</th>\n",
       "      <th>Regime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>285496.58</td>\n",
       "      <td>-0.47</td>\n",
       "      <td>1.25</td>\n",
       "      <td>-0.56</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>286101.17</td>\n",
       "      <td>-0.28</td>\n",
       "      <td>1.77</td>\n",
       "      <td>-0.39</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>287684.78</td>\n",
       "      <td>0.03</td>\n",
       "      <td>2.04</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>289724.92</td>\n",
       "      <td>0.32</td>\n",
       "      <td>2.28</td>\n",
       "      <td>-0.27</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>295386.41</td>\n",
       "      <td>0.40</td>\n",
       "      <td>2.38</td>\n",
       "      <td>-0.22</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>721485.11</td>\n",
       "      <td>-0.17</td>\n",
       "      <td>1.20</td>\n",
       "      <td>-0.06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>717529.56</td>\n",
       "      <td>-0.30</td>\n",
       "      <td>2.39</td>\n",
       "      <td>-0.34</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>713029.46</td>\n",
       "      <td>0.13</td>\n",
       "      <td>2.38</td>\n",
       "      <td>0.09</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>716175.76</td>\n",
       "      <td>0.11</td>\n",
       "      <td>2.45</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>720036.14</td>\n",
       "      <td>0.10</td>\n",
       "      <td>2.36</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>293 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      DTCTHFNM  T1YFFM  AAAFFM  TB3SMFFM  Regime\n",
       "0    285496.58   -0.47    1.25     -0.56       0\n",
       "1    286101.17   -0.28    1.77     -0.39       0\n",
       "2    287684.78    0.03    2.04     -0.35       0\n",
       "3    289724.92    0.32    2.28     -0.27       0\n",
       "4    295386.41    0.40    2.38     -0.22       0\n",
       "..         ...     ...     ...       ...     ...\n",
       "288  721485.11   -0.17    1.20     -0.06       0\n",
       "289  717529.56   -0.30    2.39     -0.34       1\n",
       "290  713029.46    0.13    2.38      0.09       1\n",
       "291  716175.76    0.11    2.45      0.08       1\n",
       "292  720036.14    0.10    2.36      0.08       1\n",
       "\n",
       "[293 rows x 5 columns]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "4ccfaf0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "a83a3f1f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-48-f31b5b3eb251>:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_1['Distance'] = distances\n"
     ]
    }
   ],
   "source": [
    "feature_means = data_1[features].mean()\n",
    "\n",
    "# Calculating distances from the mean for each row\n",
    "distances = np.linalg.norm(data_1[features] - feature_means, axis=1)\n",
    "\n",
    "# Sorting the data based on distances\n",
    "data_1['Distance'] = distances\n",
    "data_1_sorted = data_1.sort_values(by='Distance')\n",
    "\n",
    "# Selecting the top 18% closest rows\n",
    "top_18_percent = int(0.90 * len(data_1_sorted))\n",
    "selected_rows = data_1_sorted.head(top_18_percent)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "5f960c4d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows: 263\n"
     ]
    }
   ],
   "source": [
    "print(\"Count of Selected Rows:\", len(selected_rows))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "79403679",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "04325ba2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d07717ce",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "a1cf6351",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Selected Rows from Original Data:\n",
      "        Year        RPI   INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "173  2010-07  13176.118  94.8536  139179     9.4  130427   17777  24574   \n",
      "174  2010-08  13224.463  95.1448  139438     9.5  130422   17793  24577   \n",
      "172  2010-06  13137.355  94.4397  139141     9.4  130511   17763  24549   \n",
      "93   2003-10  11620.297  94.2079  137984     6.0  130454   21691  25253   \n",
      "94   2003-11  11690.647  94.9338  138424     5.8  130474   21688  25211   \n",
      "..       ...        ...      ...     ...     ...     ...     ...    ...   \n",
      "34   1998-11  10288.061  89.1098  132280     4.4  127231   24394  25313   \n",
      "33   1998-10  10238.215  89.1677  131999     4.5  126954   24406  25262   \n",
      "32   1998-09  10218.250  88.4515  131986     4.6  126752   24419  25249   \n",
      "31   1998-08  10182.172  88.6247  131390     4.5  126544   24420  25199   \n",
      "30   1998-07  10143.124  86.8423  131329     4.5  126205   24238  25159   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "173    546     579  ...  1.5304       178.5    181.9   217.605  1101.599976   \n",
      "174    599     580  ...  1.5661       179.5    182.8   217.923  1049.329956   \n",
      "172    536     587  ...  1.4768       178.3    182.3   217.199  1030.709961   \n",
      "93    1967    2012  ...  1.6792       144.8    134.2   184.900  1050.709961   \n",
      "94    2083    1918  ...  1.6897       144.6    134.2   185.000  1058.199951   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "34    1660    1672  ...  1.6611       130.8    122.0   164.100  1163.630005   \n",
      "33    1715    1719  ...  1.6944       130.9    122.2   163.900  1098.670044   \n",
      "32    1582    1577  ...  1.6823       130.4    122.3   163.500  1017.010010   \n",
      "31    1614    1654  ...  1.6342       130.4    122.7   163.400   957.280029   \n",
      "30    1698    1610  ...  1.6437       130.7    123.0   163.200  1120.670044   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "173  2254.699951  35.505924       0  15.3298            2.05  \n",
      "174  2114.030029  40.395546       0  14.6024            2.05  \n",
      "172  2109.239990  39.229294       0  15.3608            2.03  \n",
      "93   1932.209961  15.185060       0  21.5574            1.62  \n",
      "94   1960.260010  17.423424       0  21.7111            1.63  \n",
      "..           ...        ...     ...      ...             ...  \n",
      "34   1949.540039  14.440726       0  30.8573            1.41  \n",
      "33   1771.390015  15.433544       0  29.1347            1.57  \n",
      "32   1693.839966  14.440726       0  26.7002            1.58  \n",
      "31   1499.250000   9.386477       0  25.1321            1.50  \n",
      "30   1872.390015  11.823355       0  29.4216            1.38  \n",
      "\n",
      "[263 rows x 45 columns]\n"
     ]
    }
   ],
   "source": [
    "# Printing the entire rows from the original data corresponding to the selected rows\n",
    "selected_rows_from_original_data_1 = data.loc[selected_rows.index]\n",
    "print(\"Selected Rows from Original Data:\")\n",
    "print(selected_rows_from_original_data_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "7dc06d82",
   "metadata": {},
   "outputs": [],
   "source": [
    "#DOMAIN2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "9c40f25a",
   "metadata": {},
   "outputs": [],
   "source": [
    "features_2 = ['USGOOD', 'USTPU', 'BUSINV', 'ISRATIO', 'M1SL', 'M2SL', 'TOTRESNS', 'BUSLOANS', 'P/E', 'Dividend Yield']  # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "# Creating a new dataset with only the selected features\n",
    "data_2 = data[features_2 + [target]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "a78ba4dd",
   "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>USGOOD</th>\n",
       "      <th>USTPU</th>\n",
       "      <th>BUSINV</th>\n",
       "      <th>ISRATIO</th>\n",
       "      <th>M1SL</th>\n",
       "      <th>M2SL</th>\n",
       "      <th>TOTRESNS</th>\n",
       "      <th>BUSLOANS</th>\n",
       "      <th>P/E</th>\n",
       "      <th>Dividend Yield</th>\n",
       "      <th>Regime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>23196</td>\n",
       "      <td>23947</td>\n",
       "      <td>990730</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1123.5</td>\n",
       "      <td>3647.9</td>\n",
       "      <td>56.9</td>\n",
       "      <td>721.8115</td>\n",
       "      <td>18.6845</td>\n",
       "      <td>2.26</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>23280</td>\n",
       "      <td>23988</td>\n",
       "      <td>991277</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1118.5</td>\n",
       "      <td>3661.8</td>\n",
       "      <td>53.8</td>\n",
       "      <td>724.1690</td>\n",
       "      <td>18.8140</td>\n",
       "      <td>2.16</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23276</td>\n",
       "      <td>24030</td>\n",
       "      <td>989146</td>\n",
       "      <td>1.46</td>\n",
       "      <td>1122.6</td>\n",
       "      <td>3687.0</td>\n",
       "      <td>54.9</td>\n",
       "      <td>727.3465</td>\n",
       "      <td>18.9630</td>\n",
       "      <td>2.18</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23316</td>\n",
       "      <td>24043</td>\n",
       "      <td>992111</td>\n",
       "      <td>1.46</td>\n",
       "      <td>1124.8</td>\n",
       "      <td>3697.8</td>\n",
       "      <td>55.9</td>\n",
       "      <td>732.0780</td>\n",
       "      <td>18.7388</td>\n",
       "      <td>2.19</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>23358</td>\n",
       "      <td>24137</td>\n",
       "      <td>993761</td>\n",
       "      <td>1.45</td>\n",
       "      <td>1116.5</td>\n",
       "      <td>3709.6</td>\n",
       "      <td>53.2</td>\n",
       "      <td>736.4489</td>\n",
       "      <td>19.1670</td>\n",
       "      <td>2.15</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>21205</td>\n",
       "      <td>27830</td>\n",
       "      <td>2013829</td>\n",
       "      <td>1.38</td>\n",
       "      <td>4003.1</td>\n",
       "      <td>15446.9</td>\n",
       "      <td>1726.9</td>\n",
       "      <td>2348.3682</td>\n",
       "      <td>25.3952</td>\n",
       "      <td>1.97</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>21086</td>\n",
       "      <td>27723</td>\n",
       "      <td>2007046</td>\n",
       "      <td>1.45</td>\n",
       "      <td>4256.7</td>\n",
       "      <td>15989.9</td>\n",
       "      <td>2099.4</td>\n",
       "      <td>2567.9365</td>\n",
       "      <td>22.2177</td>\n",
       "      <td>2.25</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>18698</td>\n",
       "      <td>24475</td>\n",
       "      <td>1979748</td>\n",
       "      <td>1.67</td>\n",
       "      <td>4799.1</td>\n",
       "      <td>17020.1</td>\n",
       "      <td>2953.6</td>\n",
       "      <td>2932.6981</td>\n",
       "      <td>25.0359</td>\n",
       "      <td>2.41</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>19374</td>\n",
       "      <td>24858</td>\n",
       "      <td>1933282</td>\n",
       "      <td>1.50</td>\n",
       "      <td>5035.3</td>\n",
       "      <td>17868.6</td>\n",
       "      <td>3217.6</td>\n",
       "      <td>3041.1640</td>\n",
       "      <td>26.1696</td>\n",
       "      <td>2.10</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>19859</td>\n",
       "      <td>25852</td>\n",
       "      <td>1912335</td>\n",
       "      <td>1.37</td>\n",
       "      <td>5215.0</td>\n",
       "      <td>18164.1</td>\n",
       "      <td>3043.6</td>\n",
       "      <td>2933.2051</td>\n",
       "      <td>26.6508</td>\n",
       "      <td>1.95</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>293 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     USGOOD  USTPU   BUSINV  ISRATIO    M1SL     M2SL  TOTRESNS   BUSLOANS  \\\n",
       "0     23196  23947   990730     1.50  1123.5   3647.9      56.9   721.8115   \n",
       "1     23280  23988   991277     1.50  1118.5   3661.8      53.8   724.1690   \n",
       "2     23276  24030   989146     1.46  1122.6   3687.0      54.9   727.3465   \n",
       "3     23316  24043   992111     1.46  1124.8   3697.8      55.9   732.0780   \n",
       "4     23358  24137   993761     1.45  1116.5   3709.6      53.2   736.4489   \n",
       "..      ...    ...      ...      ...     ...      ...       ...        ...   \n",
       "288   21205  27830  2013829     1.38  4003.1  15446.9    1726.9  2348.3682   \n",
       "289   21086  27723  2007046     1.45  4256.7  15989.9    2099.4  2567.9365   \n",
       "290   18698  24475  1979748     1.67  4799.1  17020.1    2953.6  2932.6981   \n",
       "291   19374  24858  1933282     1.50  5035.3  17868.6    3217.6  3041.1640   \n",
       "292   19859  25852  1912335     1.37  5215.0  18164.1    3043.6  2933.2051   \n",
       "\n",
       "         P/E  Dividend Yield  Regime  \n",
       "0    18.6845            2.26       0  \n",
       "1    18.8140            2.16       0  \n",
       "2    18.9630            2.18       0  \n",
       "3    18.7388            2.19       0  \n",
       "4    19.1670            2.15       0  \n",
       "..       ...             ...     ...  \n",
       "288  25.3952            1.97       0  \n",
       "289  22.2177            2.25       1  \n",
       "290  25.0359            2.41       1  \n",
       "291  26.1696            2.10       1  \n",
       "292  26.6508            1.95       1  \n",
       "\n",
       "[293 rows x 11 columns]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "71b0be62",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 190\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "136  2007-06  13127.988  104.5617  146063     4.6  138069   22323  26549   \n",
      "137  2007-07  13134.159  104.5227  145905     4.7  138038   22277  26560   \n",
      "155  2009-01  12971.020   91.0373  142152     7.8  134055   19889  25415   \n",
      "135  2007-05  13144.673  104.5322  145903     4.4  137993   22323  26560   \n",
      "178  2010-12  13374.156   96.0599  139301     9.3  130822   17796  24686   \n",
      "..       ...        ...       ...     ...     ...     ...     ...    ...   \n",
      "228  2015-02  15169.760  105.4425  148045     5.5  140839   19537  26657   \n",
      "42   1999-07  10457.086   91.7766  133414     4.3  129414   24474  25757   \n",
      "229  2015-03  15139.301  105.1464  148128     5.4  140910   19505  26684   \n",
      "41   1999-06  10438.054   91.1994  133378     4.3  129089   24434  25701   \n",
      "40   1999-05  10400.076   91.3520  133311     4.2  128808   24445  25649   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "136   1448    1407  ...  1.9867       166.1    170.4   207.234  1503.349976   \n",
      "137   1354    1361  ...  2.0355       167.2    171.8   207.603  1455.270020   \n",
      "155    490     545  ...  1.4462       170.8    172.4   211.933   825.880005   \n",
      "135   1415    1493  ...  1.9842       166.0    169.7   206.755  1530.619995   \n",
      "178    539     632  ...  1.5595       184.0    189.3   220.472  1257.640015   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "228    886    1127  ...  1.5329       193.5    190.0   235.342  2104.500000   \n",
      "42    1669    1682  ...  1.5751       132.7    123.4   166.700  1328.719971   \n",
      "229    960    1078  ...  1.4958       194.2    189.8   235.976  2067.889893   \n",
      "41    1559    1702  ...  1.5950       132.4    122.6   166.000  1372.709961   \n",
      "40    1611    1640  ...  1.6154       132.3    122.2   166.000  1301.839966   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime       P/E  Dividend Yield  \n",
      "136  2603.229980  23.958353       0   17.7031            1.73  \n",
      "137  2546.270020  27.114885       0   18.5149            1.74  \n",
      "155  1476.420044  31.614695       1  120.3907            3.24  \n",
      "135  2604.520020  23.883192       0   18.0243            1.72  \n",
      "178  2652.870117  46.170235       0   16.2591            1.83  \n",
      "..           ...        ...     ...       ...             ...  \n",
      "228  4963.529785  12.170645       0   21.2040            1.94  \n",
      "42   2638.489990  13.679372       0   30.2257            1.20  \n",
      "229  4900.879883  10.285907       0   20.8352            1.96  \n",
      "41   2686.120117  14.278129       0   33.4644            1.24  \n",
      "40   2470.520020  12.562754       0   31.7367            1.23  \n",
      "\n",
      "[190 rows x 45 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-54-504956757177>:15: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_2['Distance'] = distances_2\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "\n",
    "# Calculating means for each feature\n",
    "feature_means_2 = data_2[features_2].mean()\n",
    "\n",
    "# Calculating distances from the mean for each row\n",
    "distances_2 = np.linalg.norm(data_2[features_2] - feature_means_2, axis=1)\n",
    "\n",
    "# Sorting the data based on distances\n",
    "data_2['Distance'] = distances_2\n",
    "data_2_sorted = data_2.sort_values(by='Distance')\n",
    "\n",
    "# Selecting the top 65% closest rows\n",
    "top_18_percent_2 = int(0.65 * len(data_2_sorted))\n",
    "selected_rows_2 = data_2_sorted.head(top_18_percent_2)\n",
    "\n",
    "# Displaying the count of selected rows\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_2))\n",
    "\n",
    "# Printing the entire rows from the original data corresponding to the selected rows\n",
    "selected_rows_from_original_data_2 = data.loc[selected_rows_2.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_2)\n"
   ]
  },
  {
   "cell_type": "raw",
   "id": "d54d55f5",
   "metadata": {},
   "source": [
    "features_3 ['SP500', 'NASDAQ', 'BUSINV', 'GOLDBAR'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "# Creating a new dataset with only the selected features\n",
    "data_3= data[features_3+ [target]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "37214a3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Domain 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "2a0542b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "features_3 = ['SP500', 'NASDAQ', 'GOLDBAR'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "\n",
    "data_3= data[features_3+ [target]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "81ab7e30",
   "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>SP500</th>\n",
       "      <th>NASDAQ</th>\n",
       "      <th>GOLDBAR</th>\n",
       "      <th>Regime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>636.020020</td>\n",
       "      <td>1059.790039</td>\n",
       "      <td>20.754421</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>640.429993</td>\n",
       "      <td>1100.050049</td>\n",
       "      <td>21.372637</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>645.500000</td>\n",
       "      <td>1101.400024</td>\n",
       "      <td>21.460953</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>654.169983</td>\n",
       "      <td>1190.520020</td>\n",
       "      <td>21.637589</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>669.119995</td>\n",
       "      <td>1243.430054</td>\n",
       "      <td>22.255806</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>2954.219971</td>\n",
       "      <td>8567.370117</td>\n",
       "      <td>18.868309</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>2584.590088</td>\n",
       "      <td>7700.100098</td>\n",
       "      <td>18.215258</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>2912.429932</td>\n",
       "      <td>8889.549805</td>\n",
       "      <td>25.572948</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>3044.310059</td>\n",
       "      <td>9489.870117</td>\n",
       "      <td>23.862782</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>3100.290039</td>\n",
       "      <td>10058.769530</td>\n",
       "      <td>26.864592</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>293 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           SP500        NASDAQ    GOLDBAR  Regime\n",
       "0     636.020020   1059.790039  20.754421       0\n",
       "1     640.429993   1100.050049  21.372637       0\n",
       "2     645.500000   1101.400024  21.460953       0\n",
       "3     654.169983   1190.520020  21.637589       0\n",
       "4     669.119995   1243.430054  22.255806       0\n",
       "..           ...           ...        ...     ...\n",
       "288  2954.219971   8567.370117  18.868309       0\n",
       "289  2584.590088   7700.100098  18.215258       1\n",
       "290  2912.429932   8889.549805  25.572948       1\n",
       "291  3044.310059   9489.870117  23.862782       1\n",
       "292  3100.290039  10058.769530  26.864592       1\n",
       "\n",
       "[293 rows x 4 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "608f2af1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 146\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "205  2013-03  13865.456  101.8114  143316     7.5  135680   18660  25602   \n",
      "206  2013-04  13930.178  101.6364  143635     7.6  135871   18653  25637   \n",
      "57   2000-10  11173.664   95.3970  137088     3.9  132351   24638  26214   \n",
      "52   2000-05  11021.996   95.6435  136630     4.0  132105   24647  26137   \n",
      "46   1999-11  10657.501   93.4371  134309     4.1  130466   24561  25937   \n",
      "..       ...        ...       ...     ...     ...     ...     ...    ...   \n",
      "164  2009-10  12709.966   89.9797  138432    10.0  130045   17915  24537   \n",
      "109  2005-03  12080.643   99.3061  140654     5.2  133156   22066  25744   \n",
      "95   2003-12  11701.031   94.8662  138411     5.7  130588   21703  25207   \n",
      "98   2004-03  11730.978   95.2088  138453     5.8  131142   21758  25382   \n",
      "49   2000-02  10899.897   94.4557  136598     4.1  131124   24609  26086   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "205   1010     936  ...  1.5080       196.5    201.6   232.282  1569.189941   \n",
      "206    835    1012  ...  1.5311       195.3    200.5   231.797  1597.569946   \n",
      "57    1549    1577  ...  1.4506       139.5    130.7   173.900  1429.400024   \n",
      "52    1575    1543  ...  1.5090       137.0    128.2   171.200  1420.599976   \n",
      "46    1648    1672  ...  1.6205       134.9    125.4   168.400  1388.910034   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "164    534     583  ...  1.6212       174.3    175.0   216.509  1036.189941   \n",
      "109   1864    2062  ...  1.9043       153.7    150.7   193.100  1180.589966   \n",
      "95    2057    1987  ...  1.7516       145.1    135.0   185.500  1111.920044   \n",
      "98    1998    2066  ...  1.8261       146.2    138.4   187.100  1126.209961   \n",
      "49    1737    1692  ...  1.6000       136.6    127.5   170.000  1366.420044   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "205  3267.520020  26.528696       0  17.8927            2.07  \n",
      "206  3328.790039  17.785048       0  17.5654            2.07  \n",
      "57   3369.629883  10.029353       0  28.5880            1.17  \n",
      "52   3400.909912  13.426710       0  27.3613            1.18  \n",
      "46   3336.159912  13.264847       0  28.8335            1.20  \n",
      "..           ...        ...     ...      ...             ...  \n",
      "164  2045.109985  30.463829       0  20.3294            2.19  \n",
      "109  1999.229980  19.164881       0  19.5721            1.69  \n",
      "95   2003.369995  17.891039       0  22.8133            1.61  \n",
      "98   1994.219971  18.733992       0  21.6579            1.60  \n",
      "49   4696.689941  12.084039       0  26.8188            1.21  \n",
      "\n",
      "[146 rows x 45 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-58-8a3931a6a31e>:11: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_3['Distance'] = distances_3\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "feature_means_3 = data_3[features_3].mean()\n",
    "\n",
    "distances_3 = np.linalg.norm(data_3[features_3] - feature_means_3, axis=1)\n",
    "\n",
    "data_3['Distance'] = distances_3\n",
    "data_3_sorted = data_3.sort_values(by='Distance')\n",
    "\n",
    "top_10_percent_3 = int(0.50 * len(data_3_sorted))\n",
    "selected_rows_3 = data_3_sorted.head(top_10_percent_3)\n",
    "\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_3))\n",
    "\n",
    "selected_rows_from_original_data_3 = data.loc[selected_rows_3.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "cbe9b87f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 131\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "148  2008-06  13359.533  102.8445  145737     5.6  137687   21483  26323   \n",
      "106  2004-12  12354.110   98.3318  140125     5.4  132624   22005  25622   \n",
      "150  2008-08  13054.138  100.7353  145203     6.1  137213   21250  26166   \n",
      "105  2004-11  11945.706   97.6137  140231     5.4  132503   21997  25598   \n",
      "147  2008-05  13704.098  103.0863  145908     5.4  137858   21599  26384   \n",
      "..       ...        ...       ...     ...     ...     ...     ...    ...   \n",
      "165  2009-11  12743.645   90.3375  138659     9.9  130057   17869  24519   \n",
      "179  2011-01  13541.211   95.9364  139250     9.1  130841   17784  24721   \n",
      "172  2010-06  13137.355   94.4397  139141     9.4  130511   17763  24549   \n",
      "181  2011-03  13540.028   96.4643  139639     9.0  131288   17906  24845   \n",
      "178  2010-12  13374.156   96.0599  139301     9.3  130822   17796  24686   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "148   1046    1180  ...  1.9664       181.2    195.3   217.463  1280.000000   \n",
      "106   2042    2082  ...  1.9286       151.4    147.7   191.700  1211.920044   \n",
      "150    844     858  ...  1.8865       182.0    198.1   218.690  1282.829956   \n",
      "105   1782    2079  ...  1.8607       152.1    147.9   191.700  1173.819946   \n",
      "147    973     995  ...  1.9650       178.4    191.1   215.208  1400.380005   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "165    588     623  ...  1.6599       176.6    176.9   217.234  1095.630005   \n",
      "179    630     576  ...  1.5782       185.4    192.0   221.187  1286.119995   \n",
      "172    536     587  ...  1.4768       178.3    182.3   217.199  1030.709961   \n",
      "181    600     583  ...  1.6159       188.8    197.6   223.046  1325.829956   \n",
      "178    539     632  ...  1.5595       184.0    189.3   220.472  1257.640015   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "148  2292.979980  37.821716       1  24.9173            2.14  \n",
      "106  2175.439941  19.372845       0  20.6989            1.62  \n",
      "150  2367.520020  28.869198       1  27.9180            2.25  \n",
      "105  2096.810059  19.580919       0  20.0482            1.66  \n",
      "147  2522.659912  33.326710       1  27.2607            2.04  \n",
      "..           ...        ...     ...      ...             ...  \n",
      "165  2144.600098  36.195393       0  21.4956            2.10  \n",
      "179  2700.080078  41.247593       0  15.8175            1.79  \n",
      "172  2109.239990  39.229294       0  15.3608            2.03  \n",
      "181  2781.070068  45.170807       0  16.3059            1.80  \n",
      "178  2652.870117  46.170235       0  16.2591            1.83  \n",
      "\n",
      "[131 rows x 45 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-59-4378b5045070>:18: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_4['Distance'] = distances_4\n"
     ]
    }
   ],
   "source": [
    "features_4 = ['FEDFUNDS', 'TB3MS', 'TB6MS', 'GS1', 'GS5', 'GS10','AAA', 'TB3SMFFM'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "data_4 = data[features_4+ [target]]\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "\n",
    "feature_means_4 = data_4[features_4].mean()\n",
    "\n",
    "\n",
    "distances_4 = np.linalg.norm(data_4[features_4] - feature_means_4, axis=1)\n",
    "\n",
    "data_4['Distance'] = distances_4\n",
    "data_4_sorted = data_4.sort_values(by='Distance')\n",
    "\n",
    "top_10_percent_4 = int(0.45 * len(data_4_sorted))\n",
    "selected_rows_4 = data_4_sorted.head(top_10_percent_4)\n",
    "\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_4))\n",
    "\n",
    "selected_rows_from_original_data_4 = data.loc[selected_rows_4.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_4)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "4ddad6bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 117\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "131  2007-01  13033.906  102.4933  146028     4.6  137477   22439  26418   \n",
      "132  2007-02  13096.338  103.5264  146057     4.5  137558   22334  26466   \n",
      "130  2006-12  12998.472  103.0292  145970     4.4  137249   22405  26372   \n",
      "166  2009-12  12808.646   90.6132  138013     9.9  129788   17792  24409   \n",
      "167  2010-01  12860.097   91.6642  138438     9.8  129790   17707  24476   \n",
      "..       ...        ...       ...     ...     ...     ...     ...    ...   \n",
      "56   2000-09  11159.329   95.6790  136893     3.9  132353   24642  26194   \n",
      "60   2001-01  11276.396   94.5448  137778     4.2  132698   24533  26202   \n",
      "57   2000-10  11173.664   95.3970  137088     3.9  132351   24638  26214   \n",
      "103  2004-08  11944.486   96.4095  139573     5.4  131937   21944  25497   \n",
      "102  2004-07  11898.458   96.3385  139556     5.5  131848   21900  25487   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "131   1409    1626  ...  1.9587       160.9    164.1   203.437  1438.239990   \n",
      "132   1480    1598  ...  1.9589       162.7    165.3   204.226  1406.819946   \n",
      "130   1649    1638  ...  1.9629       161.1    164.7   203.100  1418.300049   \n",
      "166    581     664  ...  1.6226       177.1    177.8   217.347  1115.099976   \n",
      "167    614     636  ...  1.6158       178.9    180.8   217.488  1073.869995   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "56    1507    1570  ...  1.4336       139.0    130.3   173.600  1436.510010   \n",
      "60    1600    1699  ...  1.4775       141.7    132.1   175.600  1366.010010   \n",
      "57    1549    1577  ...  1.4506       139.5    130.7   173.900  1429.400024   \n",
      "103   2024    2056  ...  1.8203       148.6    144.4   189.200  1104.239990   \n",
      "102   2002    2112  ...  1.8438       148.2    142.9   189.100  1101.719971   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "131  2463.929932  24.284945       0  17.2969            1.76  \n",
      "132  2416.149902  24.489918       0  16.9191            1.75  \n",
      "130  2415.290039  25.170416       0  17.4003            1.76  \n",
      "166  2269.149902  33.697403       0  21.8776            2.02  \n",
      "167  2147.350098  29.795422       0  17.6247            1.98  \n",
      "..           ...        ...     ...      ...             ...  \n",
      "56   3672.820068  11.435337       0  26.7507            1.11  \n",
      "60   2772.729980  11.686364       0  30.0618            1.21  \n",
      "57   3369.629883  10.029353       0  28.5880            1.17  \n",
      "103  1838.099976  15.925924       0  19.1144            1.74  \n",
      "102  1887.359985  15.225182       0  19.0708            1.70  \n",
      "\n",
      "[117 rows x 45 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-60-d49a4746d3cb>:18: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_5['Distance'] = distances_5\n"
     ]
    }
   ],
   "source": [
    "features_5 = ['AMDMUO'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "data_5 = data[features_5 + [target]]\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "\n",
    "feature_means_5 = data_5[features_5].mean()\n",
    "\n",
    "\n",
    "distances_5 = np.linalg.norm(data_5[features_5] - feature_means_5, axis=1)\n",
    "\n",
    "data_5['Distance'] = distances_5\n",
    "data_5_sorted = data_5.sort_values(by='Distance')\n",
    "\n",
    "top_10_percent_5 = int(0.40 * len(data_5_sorted))\n",
    "selected_rows_5 = data_5_sorted.head(top_10_percent_5)\n",
    "\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_5))\n",
    "\n",
    "selected_rows_from_original_data_5 = data.loc[selected_rows_5.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "43c9e569",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 102\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "115  2005-09  12202.979   98.0779  142401     5.0  134566   22226  25988   \n",
      "114  2005-08  12241.010   99.9435  142434     4.9  134498   22228  25997   \n",
      "112  2005-06  12212.817   99.9853  141714     5.0  133942   22185  25903   \n",
      "116  2005-10  12293.246   99.3149  142548     5.0  134655   22292  25997   \n",
      "173  2010-07  13176.118   94.8536  139179     9.4  130427   17777  24574   \n",
      "..       ...        ...       ...     ...     ...     ...     ...    ...   \n",
      "161  2009-07  12774.081   88.0323  139901     9.5  130668   18278  24734   \n",
      "140  2007-10  13139.629  104.7171  145946     4.7  138174   22056  26585   \n",
      "183  2011-05  13505.003   96.3377  139624     9.0  131703   18007  24947   \n",
      "48   2000-01  10855.813   94.1758  136559     4.0  131005   24628  26061   \n",
      "253  2017-03  15841.812  103.3430  152978     4.4  145944   19964  27339   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "115   2151    2263  ...  1.8064       158.8    157.2   198.800  1228.810059   \n",
      "114   2095    2219  ...  1.7944       156.3    153.2   196.100  1220.329956   \n",
      "112   2068    2178  ...  1.8177       153.9    151.1   193.700  1191.329956   \n",
      "116   2065    2170  ...  1.7651       160.5    162.3   199.100  1207.010010   \n",
      "173    546     579  ...  1.5304       178.5    181.9   217.605  1101.599976   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "161    594     595  ...  1.6378       171.5    171.1   214.726   987.479980   \n",
      "140   1264    1192  ...  2.0449       169.3    172.7   209.190  1549.380005   \n",
      "183    561     618  ...  1.6332       191.5    202.0   224.806  1345.199951   \n",
      "48    1636    1727  ...  1.6404       135.2    126.4   169.300  1394.459961   \n",
      "253   1186    1292  ...  1.2347       197.1    188.9   243.721  2362.719971   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "115  2151.689941  23.466402       0  18.4589            1.75  \n",
      "114  2152.090088  21.180351       0  18.3315            1.74  \n",
      "112  2056.959961  20.219084       0  18.8026            1.74  \n",
      "116  2120.300049  20.396790       0  17.2603            1.82  \n",
      "173  2254.699951  35.505924       0  15.3298            2.05  \n",
      "..           ...        ...     ...      ...             ...  \n",
      "161  1978.500000  29.590523       0  78.7464            2.67  \n",
      "140  2859.120117  36.370220       0  23.4116            1.77  \n",
      "183  2835.300049  41.559589       0  16.0391            1.80  \n",
      "48   3940.350098  12.130337       0  27.3692            1.17  \n",
      "253  5911.740234  18.192139       0  23.5589            1.96  \n",
      "\n",
      "[102 rows x 45 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-61-d448735d5c25>:16: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_6['Distance'] = distances_6\n"
     ]
    }
   ],
   "source": [
    "features_6 = ['RPI','INDPRO','CE16OV','UNRATE','PAYEMS','HOUST','PERMIT','DPCERA3M086SBEA','AMTMNO'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "data_6 = data[features_6 + [target]]\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "feature_means_6 = data_6[features_6].mean()\n",
    "\n",
    "distances_6 = np.linalg.norm(data_6[features_6] - feature_means_6, axis=1)\n",
    "\n",
    "data_6['Distance'] = distances_6\n",
    "data_6_sorted = data_6.sort_values(by='Distance')\n",
    "\n",
    "top_10_percent_6 = int(0.35 * len(data_6_sorted))\n",
    "selected_rows_6 = data_6_sorted.head(top_10_percent_6)\n",
    "\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_6))\n",
    "\n",
    "selected_rows_from_original_data_6 = data.loc[selected_rows_6.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_6)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "20b59fd0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 43\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "157  2009-03  12778.190   89.0160  140707     8.7  132512   19227  25123   \n",
      "158  2009-04  12837.219   88.3056  140656     9.0  131817   18894  24942   \n",
      "134  2007-04  13151.498  104.4814  145586     4.5  137842   22350  26527   \n",
      "159  2009-05  13027.964   87.4155  140248     9.4  131475   18655  24893   \n",
      "133  2007-03  13147.949  103.7521  146320     4.4  137793   22391  26541   \n",
      "135  2007-05  13144.673  104.5322  145903     4.4  137993   22323  26560   \n",
      "138  2007-08  13137.613  104.7556  145682     4.6  138015   22165  26558   \n",
      "136  2007-06  13127.988  104.5617  146063     4.6  138069   22323  26549   \n",
      "132  2007-02  13096.338  103.5264  146057     4.5  137558   22334  26466   \n",
      "156  2009-02  12813.805   90.4502  141640     8.3  133312   19575  25268   \n",
      "139  2007-09  13165.169  105.1611  146244     4.7  138095   22093  26587   \n",
      "126  2006-08  12709.549  102.3327  144625     4.7  136694   22572  26208   \n",
      "161  2009-07  12774.081   88.0323  139901     9.5  130668   18278  24734   \n",
      "137  2007-07  13134.159  104.5227  145905     4.7  138038   22277  26560   \n",
      "154  2008-12  13121.721   93.2521  143369     7.3  134839   20322  25551   \n",
      "124  2006-06  12730.660  101.9693  144353     4.6  136325   22598  26168   \n",
      "140  2007-10  13139.629  104.7171  145946     4.7  138174   22056  26585   \n",
      "155  2009-01  12971.020   91.0373  142152     7.8  134055   19889  25415   \n",
      "130  2006-12  12998.472  103.0292  145970     4.4  137249   22405  26372   \n",
      "125  2006-07  12715.713  101.9319  144202     4.7  136520   22590  26191   \n",
      "131  2007-01  13033.906  102.4933  146028     4.6  137477   22439  26418   \n",
      "123  2006-05  12701.954  101.5790  144089     4.6  136244   22597  26149   \n",
      "160  2009-06  12825.058   87.0742  140009     9.5  131008   18422  24835   \n",
      "122  2006-04  12708.846  101.6813  143761     4.7  136205   22631  26153   \n",
      "127  2006-09  12795.596  102.1252  144815     4.5  136843   22537  26242   \n",
      "119  2006-01  12615.534  101.0627  143150     4.7  135429   22467  26092   \n",
      "116  2005-10  12293.246   99.3149  142548     5.0  134655   22292  25997   \n",
      "129  2006-11  12933.502  101.9688  145534     4.5  137063   22408  26321   \n",
      "121  2006-03  12704.101  101.2751  143741     4.7  136047   22572  26169   \n",
      "162  2009-08  12743.854   89.0190  139492     9.6  130485   18151  24697   \n",
      "128  2006-10  12870.822  102.0668  145314     4.4  136852   22456  26268   \n",
      "163  2009-09  12763.787   89.6926  138818     9.8  130244   18043  24643   \n",
      "118  2005-12  12463.963  100.9437  142752     4.9  135149   22376  26055   \n",
      "120  2006-02  12671.717  101.0669  143457     4.8  135737   22535  26128   \n",
      "117  2005-11  12396.530  100.3216  142499     5.0  134993   22357  26036   \n",
      "164  2009-10  12709.966   89.9797  138432    10.0  130045   17915  24537   \n",
      "142  2007-12  13177.454  105.3457  146273     5.0  138392   21976  26640   \n",
      "141  2007-11  13127.498  105.3338  146595     4.7  138284   22015  26643   \n",
      "115  2005-09  12202.979   98.0779  142401     5.0  134566   22226  25988   \n",
      "165  2009-11  12743.645   90.3375  138659     9.9  130057   17869  24519   \n",
      "143  2008-01  13188.347  105.0619  146378     5.0  138403   21947  26630   \n",
      "153  2008-11  13211.827   96.0605  144100     6.8  135545   20623  25731   \n",
      "166  2009-12  12808.646   90.6132  138013     9.9  129788   17792  24409   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "157    505     513  ...  1.4170       169.1    168.3   212.495   797.869995   \n",
      "158    478     521  ...  1.4712       170.0    168.4   212.709   872.809998   \n",
      "134   1490    1470  ...  1.9879       165.3    168.3   205.904  1482.369995   \n",
      "159    540     556  ...  1.5418       170.3    169.4   213.022   919.140015   \n",
      "133   1495    1596  ...  1.9474       164.1    166.7   205.288  1420.859985   \n",
      "135   1415    1493  ...  1.9842       166.0    169.7   206.755  1530.619995   \n",
      "138   1330    1321  ...  2.0110       166.0    170.1   207.667  1473.989990   \n",
      "136   1448    1407  ...  1.9867       166.1    170.4   207.234  1503.349976   \n",
      "132   1480    1598  ...  1.9589       162.7    165.3   204.226  1406.819946   \n",
      "156    582     558  ...  1.4422       170.6    170.9   212.705   735.090027   \n",
      "139   1183    1261  ...  2.0184       167.6    171.4   208.547  1526.750000   \n",
      "126   1650    1722  ...  1.8941       162.1    166.5   203.800  1303.819946   \n",
      "161    594     595  ...  1.6378       171.5    171.1   214.726   987.479980   \n",
      "137   1354    1361  ...  2.0355       167.2    171.8   207.603  1455.270020   \n",
      "154    560     554  ...  1.4854       169.7    172.3   211.398   903.250000   \n",
      "124   1802    1867  ...  1.8435       161.4    165.2   201.800  1270.199951   \n",
      "140   1264    1192  ...  2.0449       169.3    172.7   209.190  1549.380005   \n",
      "155    490     545  ...  1.4462       170.8    172.4   211.933   825.880005   \n",
      "130   1649    1638  ...  1.9629       161.1    164.7   203.100  1418.300049   \n",
      "125   1737    1763  ...  1.8443       161.0    165.5   202.900  1276.660034   \n",
      "131   1409    1626  ...  1.9587       160.9    164.1   203.437  1438.239990   \n",
      "123   1942    1905  ...  1.8687       160.6    164.4   201.300  1270.089966   \n",
      "160    585     601  ...  1.6369       173.5    171.7   214.790   919.320007   \n",
      "122   1821    1998  ...  1.7680       160.6    163.0   200.700  1310.609985   \n",
      "127   1720    1655  ...  1.8839       160.2    164.7   202.800  1335.849976   \n",
      "119   2273    2212  ...  1.7686       160.5    162.4   199.300  1280.079956   \n",
      "116   2065    2170  ...  1.7651       160.5    162.3   199.100  1207.010010   \n",
      "129   1570    1535  ...  1.9125       160.0    163.8   202.000  1400.630005   \n",
      "121   1969    2118  ...  1.7442       159.3    161.5   199.700  1294.869995   \n",
      "162    586     616  ...  1.6532       173.9    173.8   215.445  1020.619995   \n",
      "128   1491    1570  ...  1.8765       158.7    163.0   201.900  1377.939941   \n",
      "163    585     609  ...  1.6323       173.5    174.3   215.861  1057.079956   \n",
      "118   1994    2120  ...  1.7458       159.6    160.2   198.100  1248.290039   \n",
      "120   2119    2141  ...  1.7480       158.7    161.6   199.400  1280.660034   \n",
      "117   2147    2218  ...  1.7349       158.7    160.3   198.100  1249.479980   \n",
      "164    534     583  ...  1.6212       174.3    175.0   216.509  1036.189941   \n",
      "142   1037    1149  ...  2.0161       171.7    176.9   211.445  1468.359985   \n",
      "141   1197    1224  ...  2.0701       172.4    176.7   210.834  1481.140015   \n",
      "115   2151    2263  ...  1.8064       158.8    157.2   198.800  1228.810059   \n",
      "165    588     623  ...  1.6599       176.6    176.9   217.234  1095.630005   \n",
      "143   1084    1094  ...  1.9702       173.3    179.2   212.174  1378.550049   \n",
      "153    652     626  ...  1.5327       172.9    179.7   213.153   896.239990   \n",
      "166    581     664  ...  1.6226       177.1    177.8   217.347  1115.099976   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime       P/E  Dividend Yield  \n",
      "157  1528.589966  27.339228       1  116.3076            3.60  \n",
      "158  1717.300049  24.539543       1  116.2197            3.15  \n",
      "134  2525.090088  23.046925       0   17.4561            1.76  \n",
      "159  1774.329956  32.112217       1  122.3888            2.90  \n",
      "133  2421.639893  23.407665       0   17.0879            1.81  \n",
      "135  2604.520020  23.883192       0   18.0243            1.72  \n",
      "138  2596.360107  26.801702       0   18.7531            1.84  \n",
      "136  2603.229980  23.958353       0   17.7031            1.73  \n",
      "132  2416.149902  24.489918       0   16.9191            1.75  \n",
      "156  1377.839966  25.467152       1  107.1560            3.43  \n",
      "139  2701.500000  33.197205       0   19.4243            1.80  \n",
      "126  2183.750000  27.348194       0   16.5944            1.85  \n",
      "161  1978.500000  29.590523       0   78.7464            2.67  \n",
      "137  2546.270020  27.114885       0   18.5149            1.74  \n",
      "154  1577.030029  31.007521       1   60.7023            3.23  \n",
      "124  2172.090088  24.178808       0   17.0520            1.87  \n",
      "140  2859.120117  36.370220       0   23.4116            1.77  \n",
      "155  1476.420044  31.614695       1  120.3907            3.24  \n",
      "130  2415.290039  25.170416       0   17.4003            1.76  \n",
      "125  2091.469971  25.159029       0   16.2487            1.88  \n",
      "131  2463.929932  24.284945       0   17.2969            1.76  \n",
      "123  2178.879883  24.826130       0   17.0505            1.80  \n",
      "160  1835.040039  28.445910       0  122.4128            2.76  \n",
      "122  2322.570068  24.720688       0   17.5944            1.77  \n",
      "127  2258.429932  25.093681       0   17.0020            1.83  \n",
      "119  2305.820068  25.515509       0   17.6150            1.75  \n",
      "116  2120.300049  20.396790       0   17.2603            1.82  \n",
      "129  2431.770020  25.681820       0   17.1835            1.77  \n",
      "121  2339.790039  22.092899       0   17.8185            1.76  \n",
      "162  2009.060059  29.420959       0   81.3892            2.42  \n",
      "128  2366.709961  25.322407       0   16.9052            1.79  \n",
      "163  2122.419922  32.134132       0   84.2967            2.29  \n",
      "118  2205.320068  22.603855       0   17.8506            1.76  \n",
      "120  2281.389893  22.198338       0   17.6230            1.77  \n",
      "117  2232.820068  21.495392       0   17.8676            1.78  \n",
      "164  2045.109985  30.463829       0   20.3294            2.19  \n",
      "142  2652.280029  34.782528       1   22.1874            1.87  \n",
      "141  2660.959961  33.386761       0   22.3805            1.88  \n",
      "115  2151.689941  23.466402       0   18.4589            1.75  \n",
      "165  2144.600098  36.195393       0   21.4956            2.10  \n",
      "143  2389.860107  42.549664       1   22.8275            2.03  \n",
      "153  1535.569946  24.488522       1   60.2312            3.23  \n",
      "166  2269.149902  33.697403       0   21.8776            2.02  \n",
      "\n",
      "[43 rows x 45 columns]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-62-abc172fe92cc>:17: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_7['Distance'] = distances_7\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "features_7 = ['EXSZUS','EXCAUS','EXUSUK','WPSFD49207','WPSID61'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "data_7 = data[features_7 + [target]]\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "\n",
    "feature_means_7 = data_7[features_7].mean()\n",
    "\n",
    "distances_7 = np.linalg.norm(data_7[features_7] - feature_means_7, axis=1)\n",
    "\n",
    "data_7['Distance'] = distances_7\n",
    "data_7_sorted = data_7.sort_values(by='Distance')\n",
    "\n",
    "top_10_percent_7 = int(0.15 * len(data_7_sorted))\n",
    "selected_rows_7 = data_7_sorted.head(top_10_percent_7)\n",
    "\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_7))\n",
    "\n",
    "selected_rows_from_original_data_7 = data.loc[selected_rows_7.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_7)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "e810b1a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count of Selected Rows (Feature Set 2): 29\n",
      "Selected Rows from Original Data (Feature Set 2):\n",
      "        Year        RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "131  2007-01  13033.906  102.4933  146028     4.6  137477   22439  26418   \n",
      "118  2005-12  12463.963  100.9437  142752     4.9  135149   22376  26055   \n",
      "120  2006-02  12671.717  101.0669  143457     4.8  135737   22535  26128   \n",
      "116  2005-10  12293.246   99.3149  142548     5.0  134655   22292  25997   \n",
      "165  2009-11  12743.645   90.3375  138659     9.9  130057   17869  24519   \n",
      "117  2005-11  12396.530  100.3216  142499     5.0  134993   22357  26036   \n",
      "125  2006-07  12715.713  101.9319  144202     4.7  136520   22590  26191   \n",
      "122  2006-04  12708.846  101.6813  143761     4.7  136205   22631  26153   \n",
      "114  2005-08  12241.010   99.9435  142434     4.9  134498   22228  25997   \n",
      "291  2020-05  18296.406   92.0613  137242    13.3  133028   19374  24858   \n",
      "168  2010-02  12852.303   91.9940  138581     9.8  129698   17627  24461   \n",
      "164  2009-10  12709.966   89.9797  138432    10.0  130045   17915  24537   \n",
      "121  2006-03  12704.101  101.2751  143741     4.7  136047   22572  26169   \n",
      "129  2006-11  12933.502  101.9688  145534     4.5  137063   22408  26321   \n",
      "126  2006-08  12709.549  102.3327  144625     4.7  136694   22572  26208   \n",
      "115  2005-09  12202.979   98.0779  142401     5.0  134566   22226  25988   \n",
      "171  2010-05  13107.196   94.2997  139241     9.6  130650   17742  24526   \n",
      "166  2009-12  12808.646   90.6132  138013     9.9  129788   17792  24409   \n",
      "153  2008-11  13211.827   96.0605  144100     6.8  135545   20623  25731   \n",
      "132  2007-02  13096.338  103.5264  146057     4.5  137558   22334  26466   \n",
      "119  2006-01  12615.534  101.0627  143150     4.7  135429   22467  26092   \n",
      "172  2010-06  13137.355   94.4397  139141     9.4  130511   17763  24549   \n",
      "127  2006-09  12795.596  102.1252  144815     4.5  136843   22537  26242   \n",
      "130  2006-12  12998.472  103.0292  145970     4.4  137249   22405  26372   \n",
      "124  2006-06  12730.660  101.9693  144353     4.6  136325   22598  26168   \n",
      "173  2010-07  13176.118   94.8536  139179     9.4  130427   17777  24574   \n",
      "128  2006-10  12870.822  102.0668  145314     4.4  136852   22456  26268   \n",
      "167  2010-01  12860.097   91.6642  138438     9.8  129790   17707  24476   \n",
      "170  2010-04  12998.264   92.9436  139297     9.9  130110   17729  24507   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "131   1409    1626  ...  1.9587       160.9    164.1   203.437  1438.239990   \n",
      "118   1994    2120  ...  1.7458       159.6    160.2   198.100  1248.290039   \n",
      "120   2119    2141  ...  1.7480       158.7    161.6   199.400  1280.660034   \n",
      "116   2065    2170  ...  1.7651       160.5    162.3   199.100  1207.010010   \n",
      "165    588     623  ...  1.6599       176.6    176.9   217.234  1095.630005   \n",
      "117   2147    2218  ...  1.7349       158.7    160.3   198.100  1249.479980   \n",
      "125   1737    1763  ...  1.8443       161.0    165.5   202.900  1276.660034   \n",
      "122   1821    1998  ...  1.7680       160.6    163.0   200.700  1310.609985   \n",
      "114   2095    2219  ...  1.7944       156.3    153.2   196.100  1220.329956   \n",
      "291   1038    1216  ...  1.2302       200.4    185.3   255.768  3044.310059   \n",
      "168    604     650  ...  1.5618       177.7    180.4   217.281  1104.489990   \n",
      "164    534     583  ...  1.6212       174.3    175.0   216.509  1036.189941   \n",
      "121   1969    2118  ...  1.7442       159.3    161.5   199.700  1294.869995   \n",
      "129   1570    1535  ...  1.9125       160.0    163.8   202.000  1400.630005   \n",
      "126   1650    1722  ...  1.8941       162.1    166.5   203.800  1303.819946   \n",
      "115   2151    2263  ...  1.8064       158.8    157.2   198.800  1228.810059   \n",
      "171    583     575  ...  1.4669       178.9    183.4   217.290  1089.410034   \n",
      "166    581     664  ...  1.6226       177.1    177.8   217.347  1115.099976   \n",
      "153    652     626  ...  1.5327       172.9    179.7   213.153   896.239990   \n",
      "132   1480    1598  ...  1.9589       162.7    165.3   204.226  1406.819946   \n",
      "119   2273    2212  ...  1.7686       160.5    162.4   199.300  1280.079956   \n",
      "172    536     587  ...  1.4768       178.3    182.3   217.199  1030.709961   \n",
      "127   1720    1655  ...  1.8839       160.2    164.7   202.800  1335.849976   \n",
      "130   1649    1638  ...  1.9629       161.1    164.7   203.100  1418.300049   \n",
      "124   1802    1867  ...  1.8435       161.4    165.2   201.800  1270.199951   \n",
      "173    546     579  ...  1.5304       178.5    181.9   217.605  1101.599976   \n",
      "128   1491    1570  ...  1.8765       158.7    163.0   201.900  1377.939941   \n",
      "167    614     636  ...  1.6158       178.9    180.8   217.488  1073.869995   \n",
      "170    687     637  ...  1.5332       178.9    182.7   217.403  1186.689941   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "131  2463.929932  24.284945       0  17.2969            1.76  \n",
      "118  2205.320068  22.603855       0  17.8506            1.76  \n",
      "120  2281.389893  22.198338       0  17.6230            1.77  \n",
      "116  2120.300049  20.396790       0  17.2603            1.82  \n",
      "165  2144.600098  36.195393       0  21.4956            2.10  \n",
      "117  2232.820068  21.495392       0  17.8676            1.78  \n",
      "125  2091.469971  25.159029       0  16.2487            1.88  \n",
      "122  2322.570068  24.720688       0  17.5944            1.77  \n",
      "114  2152.090088  21.180351       0  18.3315            1.74  \n",
      "291  9489.870117  23.862782       1  26.1696            2.10  \n",
      "168  2238.260010  32.225605       0  18.1272            2.03  \n",
      "164  2045.109985  30.463829       0  20.3294            2.19  \n",
      "121  2339.790039  22.092899       0  17.8185            1.76  \n",
      "129  2431.770020  25.681820       0  17.1835            1.77  \n",
      "126  2183.750000  27.348194       0  16.5944            1.85  \n",
      "115  2151.689941  23.466402       0  18.4589            1.75  \n",
      "171  2257.040039  36.007793       0  16.2356            1.95  \n",
      "166  2269.149902  33.697403       0  21.8776            2.02  \n",
      "153  1535.569946  24.488522       1  60.2312            3.23  \n",
      "132  2416.149902  24.489918       0  16.9191            1.75  \n",
      "119  2305.820068  25.515509       0  17.6150            1.75  \n",
      "172  2109.239990  39.229294       0  15.3608            2.03  \n",
      "127  2258.429932  25.093681       0  17.0020            1.83  \n",
      "130  2415.290039  25.170416       0  17.4003            1.76  \n",
      "124  2172.090088  24.178808       0  17.0520            1.87  \n",
      "173  2254.699951  35.505924       0  15.3298            2.05  \n",
      "128  2366.709961  25.322407       0  16.9052            1.79  \n",
      "167  2147.350098  29.795422       0  17.6247            1.98  \n",
      "170  2461.189941  37.265656       0  17.6854            1.83  \n",
      "\n",
      "[29 rows x 45 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-63-948099323422>:16: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  data_8['Distance'] = distances_8\n"
     ]
    }
   ],
   "source": [
    "features_8 = ['CPIAUCSL','ACOGNO'] # Replace these with your actual feature names\n",
    "target = 'Regime'\n",
    "\n",
    "data_8 = data[features_8 + [target]]\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import pairwise_distances_argmin_min\n",
    "\n",
    "feature_means_8 = data_8[features_8].mean()\n",
    "\n",
    "distances_8 = np.linalg.norm(data_8[features_8] - feature_means_8, axis=1)\n",
    "\n",
    "data_8['Distance'] = distances_8\n",
    "data_8_sorted = data_8.sort_values(by='Distance')\n",
    "\n",
    "top_10_percent_8 = int(0.10 * len(data_8_sorted))\n",
    "selected_rows_8 = data_8_sorted.head(top_10_percent_8)\n",
    "\n",
    "print(\"Count of Selected Rows (Feature Set 2):\", len(selected_rows_8))\n",
    "\n",
    "selected_rows_from_original_data_8 = data.loc[selected_rows_8.index]\n",
    "print(\"Selected Rows from Original Data (Feature Set 2):\")\n",
    "print(selected_rows_from_original_data_8)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "4b1b2375",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        Year        RPI   INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  \\\n",
      "0    2010-07  13176.118  94.8536  139179     9.4  130427   17777  24574   \n",
      "1    2010-08  13224.463  95.1448  139438     9.5  130422   17793  24577   \n",
      "2    2010-06  13137.355  94.4397  139141     9.4  130511   17763  24549   \n",
      "3    2003-10  11620.297  94.2079  137984     6.0  130454   21691  25253   \n",
      "4    2003-11  11690.647  94.9338  138424     5.8  130474   21688  25211   \n",
      "..       ...        ...      ...     ...     ...     ...     ...    ...   \n",
      "258  1998-11  10288.061  89.1098  132280     4.4  127231   24394  25313   \n",
      "259  1998-10  10238.215  89.1677  131999     4.5  126954   24406  25262   \n",
      "260  1998-09  10218.250  88.4515  131986     4.6  126752   24419  25249   \n",
      "261  1998-08  10182.172  88.6247  131390     4.5  126544   24420  25199   \n",
      "262  1998-07  10143.124  86.8423  131329     4.5  126205   24238  25159   \n",
      "\n",
      "     HOUST  PERMIT  ...  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL        SP500  \\\n",
      "0      546     579  ...  1.5304       178.5    181.9   217.605  1101.599976   \n",
      "1      599     580  ...  1.5661       179.5    182.8   217.923  1049.329956   \n",
      "2      536     587  ...  1.4768       178.3    182.3   217.199  1030.709961   \n",
      "3     1967    2012  ...  1.6792       144.8    134.2   184.900  1050.709961   \n",
      "4     2083    1918  ...  1.6897       144.6    134.2   185.000  1058.199951   \n",
      "..     ...     ...  ...     ...         ...      ...       ...          ...   \n",
      "258   1660    1672  ...  1.6611       130.8    122.0   164.100  1163.630005   \n",
      "259   1715    1719  ...  1.6944       130.9    122.2   163.900  1098.670044   \n",
      "260   1582    1577  ...  1.6823       130.4    122.3   163.500  1017.010010   \n",
      "261   1614    1654  ...  1.6342       130.4    122.7   163.400   957.280029   \n",
      "262   1698    1610  ...  1.6437       130.7    123.0   163.200  1120.670044   \n",
      "\n",
      "          NASDAQ    GOLDBAR  Regime      P/E  Dividend Yield  \n",
      "0    2254.699951  35.505924       0  15.3298            2.05  \n",
      "1    2114.030029  40.395546       0  14.6024            2.05  \n",
      "2    2109.239990  39.229294       0  15.3608            2.03  \n",
      "3    1932.209961  15.185060       0  21.5574            1.62  \n",
      "4    1960.260010  17.423424       0  21.7111            1.63  \n",
      "..           ...        ...     ...      ...             ...  \n",
      "258  1949.540039  14.440726       0  30.8573            1.41  \n",
      "259  1771.390015  15.433544       0  29.1347            1.57  \n",
      "260  1693.839966  14.440726       0  26.7002            1.58  \n",
      "261  1499.250000   9.386477       0  25.1321            1.50  \n",
      "262  1872.390015  11.823355       0  29.4216            1.38  \n",
      "\n",
      "[263 rows x 45 columns]\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "combined_data = pd.concat([selected_rows_from_original_data_1, \n",
    "                          selected_rows_from_original_data_2, \n",
    "                          selected_rows_from_original_data_3,selected_rows_from_original_data_4, \n",
    "                          selected_rows_from_original_data_5, \n",
    "                          selected_rows_from_original_data_6,selected_rows_from_original_data_7,selected_rows_from_original_data_8])\n",
    "\n",
    "unique_combined_data = combined_data.drop_duplicates()\n",
    "\n",
    "unique_combined_data = unique_combined_data.reset_index(drop=True)\n",
    "\n",
    "print(unique_combined_data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "17982ad4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Modified Dataset after eliminating outliers:\n",
      "           RPI   INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  HOUST  PERMIT  \\\n",
      "0    13176.118  94.8536  139179     9.4  130427   17777  24574    546     579   \n",
      "1    13224.463  95.1448  139438     9.5  130422   17793  24577    599     580   \n",
      "2    13137.355  94.4397  139141     9.4  130511   17763  24549    536     587   \n",
      "6    13107.196  94.2997  139241     9.6  130650   17742  24526    583     575   \n",
      "7    13222.624  95.3637  139396     9.5  130357   17787  24623    594     563   \n",
      "..         ...      ...     ...     ...     ...     ...    ...    ...     ...   \n",
      "258  10288.061  89.1098  132280     4.4  127231   24394  25313   1660    1672   \n",
      "259  10238.215  89.1677  131999     4.5  126954   24406  25262   1715    1719   \n",
      "260  10218.250  88.4515  131986     4.6  126752   24419  25249   1582    1577   \n",
      "261  10182.172  88.6247  131390     4.5  126544   24420  25199   1614    1654   \n",
      "262  10143.124  86.8423  131329     4.5  126205   24238  25159   1698    1610   \n",
      "\n",
      "     DPCERA3M086SBEA  ...  EXCAUS  EXUSUK  WPSFD49207  WPSID61  CPIAUCSL  \\\n",
      "0             96.836  ...  1.0422  1.5304       178.5    181.9   217.605   \n",
      "1             97.120  ...  1.0404  1.5661       179.5    182.8   217.923   \n",
      "2             96.603  ...  1.0376  1.4768       178.3    182.3   217.199   \n",
      "6             96.355  ...  1.0403  1.4669       178.9    183.4   217.290   \n",
      "7             97.227  ...  1.0330  1.5591       180.1    183.5   218.275   \n",
      "..               ...  ...     ...     ...         ...      ...       ...   \n",
      "258           72.399  ...  1.5404  1.6611       130.8    122.0   164.100   \n",
      "259           72.203  ...  1.5452  1.6944       130.9    122.2   163.900   \n",
      "260           71.960  ...  1.5218  1.6823       130.4    122.3   163.500   \n",
      "261           71.429  ...  1.5346  1.6342       130.4    122.7   163.400   \n",
      "262           71.104  ...  1.4869  1.6437       130.7    123.0   163.200   \n",
      "\n",
      "           SP500       NASDAQ    GOLDBAR      P/E  Dividend Yield  \n",
      "0    1101.599976  2254.699951  35.505924  15.3298            2.05  \n",
      "1    1049.329956  2114.030029  40.395546  14.6024            2.05  \n",
      "2    1030.709961  2109.239990  39.229294  15.3608            2.03  \n",
      "6    1089.410034  2257.040039  36.007793  16.2356            1.95  \n",
      "7    1141.199951  2368.620117  40.093815  15.8809            1.99  \n",
      "..           ...          ...        ...      ...             ...  \n",
      "258  1163.630005  1949.540039  14.440726  30.8573            1.41  \n",
      "259  1098.670044  1771.390015  15.433544  29.1347            1.57  \n",
      "260  1017.010010  1693.839966  14.440726  26.7002            1.58  \n",
      "261   957.280029  1499.250000   9.386477  25.1321            1.50  \n",
      "262  1120.670044  1872.390015  11.823355  29.4216            1.38  \n",
      "\n",
      "[249 rows x 43 columns]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import LocalOutlierFactor\n",
    "unique_combined_data = unique_combined_data.drop('Year', axis=1)\n",
    "X = unique_combined_data.drop('Regime', axis=1)\n",
    "y = unique_combined_data['Regime']\n",
    "lof = LocalOutlierFactor(contamination=0.05)  \n",
    "outliers = lof.fit_predict(X)\n",
    "\n",
    "# Filter out outliers from the dataset\n",
    "X_no_outliers = X[outliers != -1]\n",
    "y_no_outliers = y[outliers != -1]\n",
    "\n",
    "print(\"\\nModified Dataset after eliminating outliers:\")\n",
    "print(X_no_outliers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "bb24de6c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Modified Dataset after upsampling:\n",
      "           RPI    INDPRO  CE16OV  UNRATE  PAYEMS  USGOOD  USTPU  HOUST  \\\n",
      "0    13176.118   94.8536  139179     9.4  130427   17777  24574    546   \n",
      "1    13224.463   95.1448  139438     9.5  130422   17793  24577    599   \n",
      "2    13137.355   94.4397  139141     9.4  130511   17763  24549    536   \n",
      "6    13107.196   94.2997  139241     9.6  130650   17742  24526    583   \n",
      "7    13222.624   95.3637  139396     9.5  130357   17787  24623    594   \n",
      "..         ...       ...     ...     ...     ...     ...    ...    ...   \n",
      "235  13188.347  105.0619  146378     5.0  138403   21947  26630   1084   \n",
      "181  11349.539   93.7201  137783     4.3  132747   24409  26158   1590   \n",
      "237  13177.454  105.3457  146273     5.0  138392   21976  26640   1037   \n",
      "223  13359.533  102.8445  145737     5.6  137687   21483  26323   1046   \n",
      "133  11278.262   90.9315  136392     5.3  131454   23378  25744   1540   \n",
      "\n",
      "     PERMIT  DPCERA3M086SBEA  ...  EXCAUS  EXUSUK  WPSFD49207  WPSID61  \\\n",
      "0       579           96.836  ...  1.0422  1.5304       178.5    181.9   \n",
      "1       580           97.120  ...  1.0404  1.5661       179.5    182.8   \n",
      "2       587           96.603  ...  1.0376  1.4768       178.3    182.3   \n",
      "6       575           96.355  ...  1.0403  1.4669       178.9    183.4   \n",
      "7       563           97.227  ...  1.0330  1.5591       180.1    183.5   \n",
      "..      ...              ...  ...     ...     ...         ...      ...   \n",
      "235    1094           96.902  ...  1.0099  1.9702       173.3    179.2   \n",
      "181    1659           79.897  ...  1.5587  1.4445       141.2    131.0   \n",
      "237    1149           96.949  ...  1.0021  2.0161       171.7    176.9   \n",
      "223    1180           96.791  ...  1.0166  1.9664       181.2    195.3   \n",
      "133    1566           81.913  ...  1.5717  1.4501       139.0    127.6   \n",
      "\n",
      "     CPIAUCSL        SP500       NASDAQ    GOLDBAR      P/E  Dividend Yield  \n",
      "0     217.605  1101.599976  2254.699951  35.505924  15.3298            2.05  \n",
      "1     217.923  1049.329956  2114.030029  40.395546  14.6024            2.05  \n",
      "2     217.199  1030.709961  2109.239990  39.229294  15.3608            2.03  \n",
      "6     217.290  1089.410034  2257.040039  36.007793  16.2356            1.95  \n",
      "7     218.275  1141.199951  2368.620117  40.093815  15.8809            1.99  \n",
      "..        ...          ...          ...        ...      ...             ...  \n",
      "235   212.174  1378.550049  2389.860107  42.549664  22.8275            2.03  \n",
      "181   176.100  1160.329956  1840.260010  10.794967  25.5354            1.35  \n",
      "237   211.445  1468.359985  2652.280029  34.782528  22.1874            1.87  \n",
      "223   217.463  1280.000000  2292.979980  37.821716  24.9173            2.14  \n",
      "133   177.600  1059.780029  1690.199951  11.851214  42.9235            1.46  \n",
      "\n",
      "[450 rows x 43 columns]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.utils import resample\n",
    "\n",
    "\n",
    "# Separate the majority and minority classes\n",
    "majority_class = X_no_outliers[y_no_outliers == 0]\n",
    "minority_class = X_no_outliers[y_no_outliers == 1]\n",
    "\n",
    "# Upsample the minority class to match the majority class\n",
    "minority_class_upsampled = resample(minority_class, replace=True, n_samples=len(majority_class), random_state=42)\n",
    "\n",
    "X_upsampled = pd.concat([majority_class, minority_class_upsampled])\n",
    "y_upsampled = pd.Series([0] * len(majority_class) + [1] * len(minority_class_upsampled))\n",
    "\n",
    "print(\"\\nModified Dataset after upsampling:\")\n",
    "print(X_upsampled)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5fe4b0ed",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "40133ba6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop specified columns\n",
    "columns_to_drop = ['PAYEMS', 'DPCERA3M086SBEA', 'AMTMTI', 'AMTMNO', 'ACOGNO', 'AMDMUO', 'BUSINV', 'ISRATIO', 'FEDFUNDS', 'EXSZUS', 'EXCAUS', 'EXUSUK', 'GOLDBAR']\n",
    "X_upsampled = X_upsampled.drop(columns=columns_to_drop)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "4bf4ad68",
   "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>RPI</th>\n",
       "      <th>INDPRO</th>\n",
       "      <th>CE16OV</th>\n",
       "      <th>UNRATE</th>\n",
       "      <th>USGOOD</th>\n",
       "      <th>USTPU</th>\n",
       "      <th>HOUST</th>\n",
       "      <th>PERMIT</th>\n",
       "      <th>M1SL</th>\n",
       "      <th>M2SL</th>\n",
       "      <th>...</th>\n",
       "      <th>TB3SMFFM</th>\n",
       "      <th>T1YFFM</th>\n",
       "      <th>AAAFFM</th>\n",
       "      <th>WPSFD49207</th>\n",
       "      <th>WPSID61</th>\n",
       "      <th>CPIAUCSL</th>\n",
       "      <th>SP500</th>\n",
       "      <th>NASDAQ</th>\n",
       "      <th>P/E</th>\n",
       "      <th>Dividend Yield</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13176.118</td>\n",
       "      <td>94.8536</td>\n",
       "      <td>139179</td>\n",
       "      <td>9.4</td>\n",
       "      <td>17777</td>\n",
       "      <td>24574</td>\n",
       "      <td>546</td>\n",
       "      <td>579</td>\n",
       "      <td>1724.0</td>\n",
       "      <td>8606.2</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.02</td>\n",
       "      <td>0.11</td>\n",
       "      <td>4.54</td>\n",
       "      <td>178.5</td>\n",
       "      <td>181.9</td>\n",
       "      <td>217.605</td>\n",
       "      <td>1101.599976</td>\n",
       "      <td>2254.699951</td>\n",
       "      <td>15.3298</td>\n",
       "      <td>2.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13224.463</td>\n",
       "      <td>95.1448</td>\n",
       "      <td>139438</td>\n",
       "      <td>9.5</td>\n",
       "      <td>17793</td>\n",
       "      <td>24577</td>\n",
       "      <td>599</td>\n",
       "      <td>580</td>\n",
       "      <td>1748.7</td>\n",
       "      <td>8656.5</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>0.07</td>\n",
       "      <td>4.30</td>\n",
       "      <td>179.5</td>\n",
       "      <td>182.8</td>\n",
       "      <td>217.923</td>\n",
       "      <td>1049.329956</td>\n",
       "      <td>2114.030029</td>\n",
       "      <td>14.6024</td>\n",
       "      <td>2.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13137.355</td>\n",
       "      <td>94.4397</td>\n",
       "      <td>139141</td>\n",
       "      <td>9.4</td>\n",
       "      <td>17763</td>\n",
       "      <td>24549</td>\n",
       "      <td>536</td>\n",
       "      <td>587</td>\n",
       "      <td>1731.6</td>\n",
       "      <td>8596.6</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.06</td>\n",
       "      <td>0.14</td>\n",
       "      <td>4.70</td>\n",
       "      <td>178.3</td>\n",
       "      <td>182.3</td>\n",
       "      <td>217.199</td>\n",
       "      <td>1030.709961</td>\n",
       "      <td>2109.239990</td>\n",
       "      <td>15.3608</td>\n",
       "      <td>2.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>13107.196</td>\n",
       "      <td>94.2997</td>\n",
       "      <td>139241</td>\n",
       "      <td>9.6</td>\n",
       "      <td>17742</td>\n",
       "      <td>24526</td>\n",
       "      <td>583</td>\n",
       "      <td>575</td>\n",
       "      <td>1710.1</td>\n",
       "      <td>8577.5</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.04</td>\n",
       "      <td>0.17</td>\n",
       "      <td>4.76</td>\n",
       "      <td>178.9</td>\n",
       "      <td>183.4</td>\n",
       "      <td>217.290</td>\n",
       "      <td>1089.410034</td>\n",
       "      <td>2257.040039</td>\n",
       "      <td>16.2356</td>\n",
       "      <td>1.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>13222.624</td>\n",
       "      <td>95.3637</td>\n",
       "      <td>139396</td>\n",
       "      <td>9.5</td>\n",
       "      <td>17787</td>\n",
       "      <td>24623</td>\n",
       "      <td>594</td>\n",
       "      <td>563</td>\n",
       "      <td>1766.2</td>\n",
       "      <td>8687.7</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.04</td>\n",
       "      <td>0.07</td>\n",
       "      <td>4.34</td>\n",
       "      <td>180.1</td>\n",
       "      <td>183.5</td>\n",
       "      <td>218.275</td>\n",
       "      <td>1141.199951</td>\n",
       "      <td>2368.620117</td>\n",
       "      <td>15.8809</td>\n",
       "      <td>1.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>13188.347</td>\n",
       "      <td>105.0619</td>\n",
       "      <td>146378</td>\n",
       "      <td>5.0</td>\n",
       "      <td>21947</td>\n",
       "      <td>26630</td>\n",
       "      <td>1084</td>\n",
       "      <td>1094</td>\n",
       "      <td>1377.7</td>\n",
       "      <td>7491.8</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.19</td>\n",
       "      <td>-1.23</td>\n",
       "      <td>1.39</td>\n",
       "      <td>173.3</td>\n",
       "      <td>179.2</td>\n",
       "      <td>212.174</td>\n",
       "      <td>1378.550049</td>\n",
       "      <td>2389.860107</td>\n",
       "      <td>22.8275</td>\n",
       "      <td>2.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>11349.539</td>\n",
       "      <td>93.7201</td>\n",
       "      <td>137783</td>\n",
       "      <td>4.3</td>\n",
       "      <td>24409</td>\n",
       "      <td>26158</td>\n",
       "      <td>1590</td>\n",
       "      <td>1659</td>\n",
       "      <td>1108.9</td>\n",
       "      <td>5060.4</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.89</td>\n",
       "      <td>-1.01</td>\n",
       "      <td>1.67</td>\n",
       "      <td>141.2</td>\n",
       "      <td>131.0</td>\n",
       "      <td>176.100</td>\n",
       "      <td>1160.329956</td>\n",
       "      <td>1840.260010</td>\n",
       "      <td>25.5354</td>\n",
       "      <td>1.35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>13177.454</td>\n",
       "      <td>105.3457</td>\n",
       "      <td>146273</td>\n",
       "      <td>5.0</td>\n",
       "      <td>21976</td>\n",
       "      <td>26640</td>\n",
       "      <td>1037</td>\n",
       "      <td>1149</td>\n",
       "      <td>1373.4</td>\n",
       "      <td>7458.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.24</td>\n",
       "      <td>-0.98</td>\n",
       "      <td>1.25</td>\n",
       "      <td>171.7</td>\n",
       "      <td>176.9</td>\n",
       "      <td>211.445</td>\n",
       "      <td>1468.359985</td>\n",
       "      <td>2652.280029</td>\n",
       "      <td>22.1874</td>\n",
       "      <td>1.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>13359.533</td>\n",
       "      <td>102.8445</td>\n",
       "      <td>145737</td>\n",
       "      <td>5.6</td>\n",
       "      <td>21483</td>\n",
       "      <td>26323</td>\n",
       "      <td>1046</td>\n",
       "      <td>1180</td>\n",
       "      <td>1404.6</td>\n",
       "      <td>7714.6</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.14</td>\n",
       "      <td>0.42</td>\n",
       "      <td>3.68</td>\n",
       "      <td>181.2</td>\n",
       "      <td>195.3</td>\n",
       "      <td>217.463</td>\n",
       "      <td>1280.000000</td>\n",
       "      <td>2292.979980</td>\n",
       "      <td>24.9173</td>\n",
       "      <td>2.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>11278.262</td>\n",
       "      <td>90.9315</td>\n",
       "      <td>136392</td>\n",
       "      <td>5.3</td>\n",
       "      <td>23378</td>\n",
       "      <td>25744</td>\n",
       "      <td>1540</td>\n",
       "      <td>1566</td>\n",
       "      <td>1165.7</td>\n",
       "      <td>5324.9</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.33</td>\n",
       "      <td>-0.16</td>\n",
       "      <td>4.54</td>\n",
       "      <td>139.0</td>\n",
       "      <td>127.6</td>\n",
       "      <td>177.600</td>\n",
       "      <td>1059.780029</td>\n",
       "      <td>1690.199951</td>\n",
       "      <td>42.9235</td>\n",
       "      <td>1.46</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>450 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           RPI    INDPRO  CE16OV  UNRATE  USGOOD  USTPU  HOUST  PERMIT  \\\n",
       "0    13176.118   94.8536  139179     9.4   17777  24574    546     579   \n",
       "1    13224.463   95.1448  139438     9.5   17793  24577    599     580   \n",
       "2    13137.355   94.4397  139141     9.4   17763  24549    536     587   \n",
       "6    13107.196   94.2997  139241     9.6   17742  24526    583     575   \n",
       "7    13222.624   95.3637  139396     9.5   17787  24623    594     563   \n",
       "..         ...       ...     ...     ...     ...    ...    ...     ...   \n",
       "235  13188.347  105.0619  146378     5.0   21947  26630   1084    1094   \n",
       "181  11349.539   93.7201  137783     4.3   24409  26158   1590    1659   \n",
       "237  13177.454  105.3457  146273     5.0   21976  26640   1037    1149   \n",
       "223  13359.533  102.8445  145737     5.6   21483  26323   1046    1180   \n",
       "133  11278.262   90.9315  136392     5.3   23378  25744   1540    1566   \n",
       "\n",
       "       M1SL    M2SL  ...  TB3SMFFM  T1YFFM  AAAFFM  WPSFD49207  WPSID61  \\\n",
       "0    1724.0  8606.2  ...     -0.02    0.11    4.54       178.5    181.9   \n",
       "1    1748.7  8656.5  ...     -0.03    0.07    4.30       179.5    182.8   \n",
       "2    1731.6  8596.6  ...     -0.06    0.14    4.70       178.3    182.3   \n",
       "6    1710.1  8577.5  ...     -0.04    0.17    4.76       178.9    183.4   \n",
       "7    1766.2  8687.7  ...     -0.04    0.07    4.34       180.1    183.5   \n",
       "..      ...     ...  ...       ...     ...     ...         ...      ...   \n",
       "235  1377.7  7491.8  ...     -1.19   -1.23    1.39       173.3    179.2   \n",
       "181  1108.9  5060.4  ...     -0.89   -1.01    1.67       141.2    131.0   \n",
       "237  1373.4  7458.0  ...     -1.24   -0.98    1.25       171.7    176.9   \n",
       "223  1404.6  7714.6  ...     -0.14    0.42    3.68       181.2    195.3   \n",
       "133  1165.7  5324.9  ...     -0.33   -0.16    4.54       139.0    127.6   \n",
       "\n",
       "     CPIAUCSL        SP500       NASDAQ      P/E  Dividend Yield  \n",
       "0     217.605  1101.599976  2254.699951  15.3298            2.05  \n",
       "1     217.923  1049.329956  2114.030029  14.6024            2.05  \n",
       "2     217.199  1030.709961  2109.239990  15.3608            2.03  \n",
       "6     217.290  1089.410034  2257.040039  16.2356            1.95  \n",
       "7     218.275  1141.199951  2368.620117  15.8809            1.99  \n",
       "..        ...          ...          ...      ...             ...  \n",
       "235   212.174  1378.550049  2389.860107  22.8275            2.03  \n",
       "181   176.100  1160.329956  1840.260010  25.5354            1.35  \n",
       "237   211.445  1468.359985  2652.280029  22.1874            1.87  \n",
       "223   217.463  1280.000000  2292.979980  24.9173            2.14  \n",
       "133   177.600  1059.780029  1690.199951  42.9235            1.46  \n",
       "\n",
       "[450 rows x 30 columns]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_upsampled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9afa915c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "2027d4f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic Regression Accuracy: 0.9805555555555555\n",
      "Decision Tree Accuracy: 0.95\n",
      "Random Forest Accuracy: 0.9861111111111112\n",
      "SVM Accuracy: 0.9444444444444444\n",
      "XGBoost Accuracy: 0.9833333333333333\n",
      "Probit Model Accuracy: 0.9638888888888889\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.svm import SVC\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.metrics import accuracy_score, roc_curve, auc\n",
    "from sklearn.model_selection import train_test_split\n",
    "from scipy.optimize import minimize\n",
    "from scipy.stats import norm\n",
    "\n",
    "# Define the probit link function\n",
    "def probit_link_function(X, beta):\n",
    "    return norm.cdf(np.dot(X, beta))\n",
    "\n",
    "# Define the log likelihood function for Probit model\n",
    "def log_likelihood(beta, X, y):\n",
    "    p = probit_link_function(X, beta)\n",
    "    return -np.sum(y * np.log(p + 1e-9) + (1 - y) * np.log(1 - p + 1e-9))\n",
    "\n",
    "\n",
    "# Step 1: Split the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_upsampled, y_upsampled, test_size=0.8, random_state=42)\n",
    "\n",
    "# Step 2: Standardize the features if necessary\n",
    "scaler = StandardScaler()\n",
    "X_train_scaled = scaler.fit_transform(X_train)\n",
    "X_test_scaled = scaler.transform(X_test)\n",
    "\n",
    "# Step 3: Train Logistic Regression Model\n",
    "logreg = LogisticRegression()\n",
    "logreg.fit(X_train_scaled, y_train)\n",
    "\n",
    "# Step 4: Train Decision Tree Model\n",
    "dt_classifier = DecisionTreeClassifier()\n",
    "dt_classifier.fit(X_train_scaled, y_train)\n",
    "\n",
    "# Step 5: Train Random Forest Model\n",
    "rf_classifier = RandomForestClassifier()\n",
    "rf_classifier.fit(X_train_scaled, y_train)\n",
    "\n",
    "\n",
    "# Step 7: Train SVM Model\n",
    "svm_classifier = SVC(probability=True)\n",
    "svm_classifier.fit(X_train_scaled, y_train)\n",
    "\n",
    "# Step 8: Train XGBoost Model\n",
    "xgb_classifier = XGBClassifier()\n",
    "xgb_classifier.fit(X_train_scaled, y_train)\n",
    "\n",
    "# Step 9: Train Probit Model\n",
    "initial_beta = np.zeros(X_train_scaled.shape[1])\n",
    "result = minimize(log_likelihood, initial_beta, args=(X_train_scaled, y_train), method='BFGS')\n",
    "optimal_beta = result.x\n",
    "\n",
    "# Step 10: Make Predictions\n",
    "y_pred_logreg = logreg.predict(X_test_scaled)\n",
    "y_pred_dt = dt_classifier.predict(X_test_scaled)\n",
    "y_pred_rf = rf_classifier.predict(X_test_scaled)\n",
    "\n",
    "y_pred_svm = svm_classifier.predict(X_test_scaled)\n",
    "y_pred_xgb = xgb_classifier.predict(X_test_scaled)\n",
    "y_pred_prob_probit = probit_link_function(X_test_scaled, optimal_beta)\n",
    "y_pred_probit = (y_pred_prob_probit >= 0.5).astype(int)\n",
    "\n",
    "# Step 11: Calculate Accuracies\n",
    "accuracy_logreg = accuracy_score(y_test, y_pred_logreg)\n",
    "accuracy_dt = accuracy_score(y_test, y_pred_dt)\n",
    "accuracy_rf = accuracy_score(y_test, y_pred_rf)\n",
    "accuracy_svm = accuracy_score(y_test, y_pred_svm)\n",
    "accuracy_xgb = accuracy_score(y_test, y_pred_xgb)\n",
    "accuracy_probit = accuracy_score(y_test, y_pred_probit)\n",
    "\n",
    "print(\"Logistic Regression Accuracy:\", accuracy_logreg)\n",
    "print(\"Decision Tree Accuracy:\", accuracy_dt)\n",
    "print(\"Random Forest Accuracy:\", accuracy_rf)\n",
    "print(\"SVM Accuracy:\", accuracy_svm)\n",
    "print(\"XGBoost Accuracy:\", accuracy_xgb)\n",
    "print(\"Probit Model Accuracy:\", accuracy_probit)\n",
    "\n",
    "# Step 12: Compute ROC curves and AUC\n",
    "fpr_logreg, tpr_logreg, _ = roc_curve(y_test, logreg.predict_proba(X_test_scaled)[:, 1])\n",
    "roc_auc_logreg = auc(fpr_logreg, tpr_logreg)\n",
    "\n",
    "fpr_dt, tpr_dt, _ = roc_curve(y_test, dt_classifier.predict_proba(X_test_scaled)[:, 1])\n",
    "roc_auc_dt = auc(fpr_dt, tpr_dt)\n",
    "\n",
    "fpr_rf, tpr_rf, _ = roc_curve(y_test, rf_classifier.predict_proba(X_test_scaled)[:, 1])\n",
    "roc_auc_rf = auc(fpr_rf, tpr_rf)\n",
    "\n",
    "\n",
    "\n",
    "fpr_svm, tpr_svm, _ = roc_curve(y_test, svm_classifier.predict_proba(X_test_scaled)[:, 1])\n",
    "roc_auc_svm = auc(fpr_svm, tpr_svm)\n",
    "\n",
    "fpr_xgb, tpr_xgb, _ = roc_curve(y_test, xgb_classifier.predict_proba(X_test_scaled)[:, 1])\n",
    "roc_auc_xgb = auc(fpr_xgb, tpr_xgb)\n",
    "\n",
    "fpr_probit, tpr_probit, _ = roc_curve(y_test, y_pred_prob_probit)\n",
    "roc_auc_probit = auc(fpr_probit, tpr_probit)\n",
    "\n",
    "# Step 13: Plot ROC curves\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(fpr_logreg, tpr_logreg, color='blue', lw=2, label='Logistic Regression (AUC = %0.2f)' % roc_auc_logreg)\n",
    "plt.plot(fpr_dt, tpr_dt, color='green', lw=2, label='Decision Tree (AUC = %0.2f)' % roc_auc_dt)\n",
    "plt.plot(fpr_rf, tpr_rf, color='red', lw=2, label='Random Forest (AUC = %0.2f)' % roc_auc_rf)\n",
    "plt.plot(fpr_svm, tpr_svm, color='purple', lw=2, label='SVM (AUC = %0.2f)' % roc_auc_svm)\n",
    "plt.plot(fpr_xgb, tpr_xgb, color='cyan', lw=2, label='XGBoost (AUC = %0.2f)' % roc_auc_xgb)\n",
    "plt.plot(fpr_probit, tpr_probit, color='magenta', lw=2, label='Probit Model (AUC = %0.2f)' % roc_auc_probit)\n",
    "plt.plot([0, 1], [0, 1], color='gray', linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curves')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "96c5228c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "50d63f34",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5407791a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "88ac5e8f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9b450201",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "36880b16",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}