Causal_Inference.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 41188 entries, 0 to 41187\n",
      "Data columns (total 21 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   age             41188 non-null  int64  \n",
      " 1   job             41188 non-null  object \n",
      " 2   marital         41188 non-null  object \n",
      " 3   education       41188 non-null  object \n",
      " 4   default         41188 non-null  object \n",
      " 5   housing         41188 non-null  object \n",
      " 6   loan            41188 non-null  object \n",
      " 7   contact         41188 non-null  object \n",
      " 8   month           41188 non-null  object \n",
      " 9   day_of_week     41188 non-null  object \n",
      " 10  duration        41188 non-null  int64  \n",
      " 11  campaign        41188 non-null  int64  \n",
      " 12  pdays           41188 non-null  int64  \n",
      " 13  previous        41188 non-null  int64  \n",
      " 14  poutcome        41188 non-null  object \n",
      " 15  emp.var.rate    41188 non-null  float64\n",
      " 16  cons.price.idx  41188 non-null  float64\n",
      " 17  cons.conf.idx   41188 non-null  float64\n",
      " 18  euribor3m       41188 non-null  float64\n",
      " 19  nr.employed     41188 non-null  float64\n",
      " 20  subscribed      41188 non-null  object \n",
      "dtypes: float64(5), int64(5), object(11)\n",
      "memory usage: 6.6+ MB\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(   age        job  marital    education  default housing loan    contact  \\\n",
       " 0   56  housemaid  married     basic.4y       no      no   no  telephone   \n",
       " 1   57   services  married  high.school  unknown      no   no  telephone   \n",
       " 2   37   services  married  high.school       no     yes   no  telephone   \n",
       " 3   40     admin.  married     basic.6y       no      no   no  telephone   \n",
       " 4   56   services  married  high.school       no      no  yes  telephone   \n",
       " \n",
       "   month day_of_week  ...  campaign  pdays  previous     poutcome emp.var.rate  \\\n",
       " 0   may         mon  ...         1    999         0  nonexistent          1.1   \n",
       " 1   may         mon  ...         1    999         0  nonexistent          1.1   \n",
       " 2   may         mon  ...         1    999         0  nonexistent          1.1   \n",
       " 3   may         mon  ...         1    999         0  nonexistent          1.1   \n",
       " 4   may         mon  ...         1    999         0  nonexistent          1.1   \n",
       " \n",
       "    cons.price.idx  cons.conf.idx  euribor3m  nr.employed  subscribed  \n",
       " 0          93.994          -36.4      4.857       5191.0          no  \n",
       " 1          93.994          -36.4      4.857       5191.0          no  \n",
       " 2          93.994          -36.4      4.857       5191.0          no  \n",
       " 3          93.994          -36.4      4.857       5191.0          no  \n",
       " 4          93.994          -36.4      4.857       5191.0          no  \n",
       " \n",
       " [5 rows x 21 columns],\n",
       " None,\n",
       "                age      duration      campaign         pdays      previous  \\\n",
       " count  41188.00000  41188.000000  41188.000000  41188.000000  41188.000000   \n",
       " mean      40.02406    258.285010      2.567593    962.475454      0.172963   \n",
       " std       10.42125    259.279249      2.770014    186.910907      0.494901   \n",
       " min       17.00000      0.000000      1.000000      0.000000      0.000000   \n",
       " 25%       32.00000    102.000000      1.000000    999.000000      0.000000   \n",
       " 50%       38.00000    180.000000      2.000000    999.000000      0.000000   \n",
       " 75%       47.00000    319.000000      3.000000    999.000000      0.000000   \n",
       " max       98.00000   4918.000000     56.000000    999.000000      7.000000   \n",
       " \n",
       "        emp.var.rate  cons.price.idx  cons.conf.idx     euribor3m   nr.employed  \n",
       " count  41188.000000    41188.000000   41188.000000  41188.000000  41188.000000  \n",
       " mean       0.081886       93.575664     -40.502600      3.621291   5167.035911  \n",
       " std        1.570960        0.578840       4.628198      1.734447     72.251528  \n",
       " min       -3.400000       92.201000     -50.800000      0.634000   4963.600000  \n",
       " 25%       -1.800000       93.075000     -42.700000      1.344000   5099.100000  \n",
       " 50%        1.100000       93.749000     -41.800000      4.857000   5191.000000  \n",
       " 75%        1.400000       93.994000     -36.400000      4.961000   5228.100000  \n",
       " max        1.400000       94.767000     -26.900000      5.045000   5228.100000  )"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Load the dataset\n",
    "file_path = 'bank_marketing_dataset.csv'\n",
    "data = pd.read_csv(file_path)\n",
    "\n",
    "# Display the first few rows and summary statistics of the dataset\n",
    "data_head = data.head()\n",
    "data_info = data.info()\n",
    "data_description = data.describe()\n",
    "\n",
    "data_head, data_info, data_description\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(   age  job  marital  education  default  housing  loan  contact  month  \\\n",
       " 0   56    4        2          1        1        1     1        1      6   \n",
       " 1   57    8        2          4        0        1     1        1      6   \n",
       " 2   37    8        2          4        1        2     1        1      6   \n",
       " 3   40    1        2          2        1        1     1        1      6   \n",
       " 4   56    8        2          4        1        1     2        1      6   \n",
       " \n",
       "    day_of_week  ...  campaign  pdays  previous  poutcome  emp.var.rate  \\\n",
       " 0            1  ...         1    999         0         1           1.1   \n",
       " 1            1  ...         1    999         0         1           1.1   \n",
       " 2            1  ...         1    999         0         1           1.1   \n",
       " 3            1  ...         1    999         0         1           1.1   \n",
       " 4            1  ...         1    999         0         1           1.1   \n",
       " \n",
       "    cons.price.idx  cons.conf.idx  euribor3m  nr.employed  subscribed  \n",
       " 0          93.994          -36.4      4.857       5191.0           0  \n",
       " 1          93.994          -36.4      4.857       5191.0           0  \n",
       " 2          93.994          -36.4      4.857       5191.0           0  \n",
       " 3          93.994          -36.4      4.857       5191.0           0  \n",
       " 4          93.994          -36.4      4.857       5191.0           0  \n",
       " \n",
       " [5 rows x 21 columns],\n",
       " age               0\n",
       " job               0\n",
       " marital           0\n",
       " education         0\n",
       " default           0\n",
       " housing           0\n",
       " loan              0\n",
       " contact           0\n",
       " month             0\n",
       " day_of_week       0\n",
       " duration          0\n",
       " campaign          0\n",
       " pdays             0\n",
       " previous          0\n",
       " poutcome          0\n",
       " emp.var.rate      0\n",
       " cons.price.idx    0\n",
       " cons.conf.idx     0\n",
       " euribor3m         0\n",
       " nr.employed       0\n",
       " subscribed        0\n",
       " dtype: int64)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "# Copying the dataset to avoid changes to the original dataframe\n",
    "data_preprocessed = data.copy()\n",
    "\n",
    "# Initialize LabelEncoder\n",
    "le = LabelEncoder()\n",
    "\n",
    "# Columns to encode\n",
    "categorical_columns = data_preprocessed.select_dtypes(include=['object']).columns.tolist()\n",
    "categorical_columns.remove('subscribed')  # We'll handle the target variable separately\n",
    "\n",
    "# Encoding categorical features\n",
    "for column in categorical_columns:\n",
    "    data_preprocessed[column] = data_preprocessed[column].replace('unknown', None)  # Replace 'unknown' with NaN\n",
    "    data_preprocessed[column] = le.fit_transform(data_preprocessed[column].astype(str))  # Encode\n",
    "\n",
    "# Encoding target variable\n",
    "data_preprocessed['subscribed'] = le.fit_transform(data_preprocessed['subscribed'])\n",
    "\n",
    "# Check the head of the dataframe and the handling of 'unknown' values\n",
    "data_preprocessed.head(), data_preprocessed.isnull().sum()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "subscribed\n",
       "0    0.887346\n",
       "1    0.112654\n",
       "Name: proportion, dtype: float64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking the distribution of the target variable 'subscribed'\n",
    "subscription_counts = data_preprocessed['subscribed'].value_counts(normalize=True)\n",
    "subscription_counts\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>job</th>\n",
       "      <th>marital</th>\n",
       "      <th>education</th>\n",
       "      <th>default</th>\n",
       "      <th>housing</th>\n",
       "      <th>loan</th>\n",
       "      <th>contact</th>\n",
       "      <th>month</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>previous</th>\n",
       "      <th>emp.var.rate</th>\n",
       "      <th>cons.price.idx</th>\n",
       "      <th>cons.conf.idx</th>\n",
       "      <th>euribor3m</th>\n",
       "      <th>nr.employed</th>\n",
       "      <th>subscribed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>999</td>\n",
       "      <td>0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>93.994</td>\n",
       "      <td>-36.4</td>\n",
       "      <td>4.857</td>\n",
       "      <td>5191.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>57</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>999</td>\n",
       "      <td>0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>93.994</td>\n",
       "      <td>-36.4</td>\n",
       "      <td>4.857</td>\n",
       "      <td>5191.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>37</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>999</td>\n",
       "      <td>0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>93.994</td>\n",
       "      <td>-36.4</td>\n",
       "      <td>4.857</td>\n",
       "      <td>5191.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>40</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>999</td>\n",
       "      <td>0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>93.994</td>\n",
       "      <td>-36.4</td>\n",
       "      <td>4.857</td>\n",
       "      <td>5191.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>56</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>999</td>\n",
       "      <td>0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>93.994</td>\n",
       "      <td>-36.4</td>\n",
       "      <td>4.857</td>\n",
       "      <td>5191.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  job  marital  education  default  housing  loan  contact  month  \\\n",
       "0   56    4        2          1        1        1     1        1      6   \n",
       "1   57    8        2          4        0        1     1        1      6   \n",
       "2   37    8        2          4        1        2     1        1      6   \n",
       "3   40    1        2          2        1        1     1        1      6   \n",
       "4   56    8        2          4        1        1     2        1      6   \n",
       "\n",
       "   campaign  pdays  previous  emp.var.rate  cons.price.idx  cons.conf.idx  \\\n",
       "0         1    999         0           1.1          93.994          -36.4   \n",
       "1         1    999         0           1.1          93.994          -36.4   \n",
       "2         1    999         0           1.1          93.994          -36.4   \n",
       "3         1    999         0           1.1          93.994          -36.4   \n",
       "4         1    999         0           1.1          93.994          -36.4   \n",
       "\n",
       "   euribor3m  nr.employed  subscribed  \n",
       "0      4.857       5191.0           0  \n",
       "1      4.857       5191.0           0  \n",
       "2      4.857       5191.0           0  \n",
       "3      4.857       5191.0           0  \n",
       "4      4.857       5191.0           0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dropping the specified columns: 'duration', 'poutcome', and 'day_of_week'\n",
    "data_preprocessed = data_preprocessed.drop(['duration', 'poutcome', 'day_of_week'], axis=1)\n",
    "\n",
    "# Display the updated DataFrame to confirm the changes\n",
    "data_preprocessed.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# Initialize the StandardScaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Columns to scale (all numeric columns except the target variable)\n",
    "numeric_columns = data_preprocessed.select_dtypes(include=['int64', 'float64']).columns.tolist()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['age',\n",
       " 'campaign',\n",
       " 'pdays',\n",
       " 'previous',\n",
       " 'emp.var.rate',\n",
       " 'cons.price.idx',\n",
       " 'cons.conf.idx',\n",
       " 'euribor3m',\n",
       " 'nr.employed']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numeric_columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Upsampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "subscribed\n",
       "0    36548\n",
       "1    36548\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.utils import resample\n",
    "\n",
    "# Separate the majority and minority classes\n",
    "data_majority = data_preprocessed[data_preprocessed.subscribed == 0]\n",
    "data_minority = data_preprocessed[data_preprocessed.subscribed == 1]\n",
    "\n",
    "# Upsample minority class\n",
    "data_minority_upsampled = resample(data_minority,\n",
    "                                   replace=True,     # sample with replacement\n",
    "                                   n_samples=len(data_majority),    # to match majority class\n",
    "                                   random_state=123) # reproducible results\n",
    "\n",
    "# Combine majority class with upsampled minority class\n",
    "data_balanced = pd.concat([data_majority, data_minority_upsampled])\n",
    "\n",
    "# Display new class counts\n",
    "data_balanced['subscribed'].value_counts()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "data_balanced=data_preprocessed.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   age  job  marital  education  default  housing  loan  contact  month  \\\n",
      "0   56    4        2          1        1        1     1        1      6   \n",
      "1   57    8        2          4        0        1     1        1      6   \n",
      "2   37    8        2          4        1        2     1        1      6   \n",
      "3   40    1        2          2        1        1     1        1      6   \n",
      "4   56    8        2          4        1        1     2        1      6   \n",
      "\n",
      "   campaign  pdays  previous  emp.var.rate  cons.price.idx  cons.conf.idx  \\\n",
      "0         1    999         0           1.1          93.994          -36.4   \n",
      "1         1    999         0           1.1          93.994          -36.4   \n",
      "2         1    999         0           1.1          93.994          -36.4   \n",
      "3         1    999         0           1.1          93.994          -36.4   \n",
      "4         1    999         0           1.1          93.994          -36.4   \n",
      "\n",
      "   euribor3m  nr.employed  subscribed  \n",
      "0      4.857       5191.0           0  \n",
      "1      4.857       5191.0           0  \n",
      "2      4.857       5191.0           0  \n",
      "3      4.857       5191.0           0  \n",
      "4      4.857       5191.0           0  \n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "# Create a copy of the data_balanced dataset\n",
    "dataset1 = data_balanced.copy()\n",
    "\n",
    "# Print the first few rows of dataset1\n",
    "print(dataset1.head())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>age</th>\n",
       "      <th>job</th>\n",
       "      <th>marital</th>\n",
       "      <th>education</th>\n",
       "      <th>default</th>\n",
       "      <th>housing</th>\n",
       "      <th>loan</th>\n",
       "      <th>contact</th>\n",
       "      <th>month</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>previous</th>\n",
       "      <th>emp.var.rate</th>\n",
       "      <th>cons.price.idx</th>\n",
       "      <th>cons.conf.idx</th>\n",
       "      <th>euribor3m</th>\n",
       "      <th>nr.employed</th>\n",
       "      <th>subscribed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.305973</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.56714</td>\n",
       "      <td>0.356658</td>\n",
       "      <td>-0.447954</td>\n",
       "      <td>0.924539</td>\n",
       "      <td>0.814609</td>\n",
       "      <td>0.71345</td>\n",
       "      <td>1.000075</td>\n",
       "      <td>0.636166</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.389478</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.56714</td>\n",
       "      <td>0.356658</td>\n",
       "      <td>-0.447954</td>\n",
       "      <td>0.924539</td>\n",
       "      <td>0.814609</td>\n",
       "      <td>0.71345</td>\n",
       "      <td>1.000075</td>\n",
       "      <td>0.636166</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.280637</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.56714</td>\n",
       "      <td>0.356658</td>\n",
       "      <td>-0.447954</td>\n",
       "      <td>0.924539</td>\n",
       "      <td>0.814609</td>\n",
       "      <td>0.71345</td>\n",
       "      <td>1.000075</td>\n",
       "      <td>0.636166</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.030120</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.56714</td>\n",
       "      <td>0.356658</td>\n",
       "      <td>-0.447954</td>\n",
       "      <td>0.924539</td>\n",
       "      <td>0.814609</td>\n",
       "      <td>0.71345</td>\n",
       "      <td>1.000075</td>\n",
       "      <td>0.636166</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.305973</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.56714</td>\n",
       "      <td>0.356658</td>\n",
       "      <td>-0.447954</td>\n",
       "      <td>0.924539</td>\n",
       "      <td>0.814609</td>\n",
       "      <td>0.71345</td>\n",
       "      <td>1.000075</td>\n",
       "      <td>0.636166</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  job  marital  education  default  housing  loan  contact  month  \\\n",
       "0  1.305973    4        2          1        1        1     1        1      6   \n",
       "1  1.389478    8        2          4        0        1     1        1      6   \n",
       "2 -0.280637    8        2          4        1        2     1        1      6   \n",
       "3 -0.030120    1        2          2        1        1     1        1      6   \n",
       "4  1.305973    8        2          4        1        1     2        1      6   \n",
       "\n",
       "   campaign     pdays  previous  emp.var.rate  cons.price.idx  cons.conf.idx  \\\n",
       "0  -0.56714  0.356658 -0.447954      0.924539        0.814609        0.71345   \n",
       "1  -0.56714  0.356658 -0.447954      0.924539        0.814609        0.71345   \n",
       "2  -0.56714  0.356658 -0.447954      0.924539        0.814609        0.71345   \n",
       "3  -0.56714  0.356658 -0.447954      0.924539        0.814609        0.71345   \n",
       "4  -0.56714  0.356658 -0.447954      0.924539        0.814609        0.71345   \n",
       "\n",
       "   euribor3m  nr.employed  subscribed  \n",
       "0   1.000075     0.636166           0  \n",
       "1   1.000075     0.636166           0  \n",
       "2   1.000075     0.636166           0  \n",
       "3   1.000075     0.636166           0  \n",
       "4   1.000075     0.636166           0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Applying standard scaling\n",
    "data_balanced[numeric_columns] = scaler.fit_transform(data_balanced[numeric_columns])\n",
    "\n",
    "# Display the scaled dataframe to confirm the changes\n",
    "data_balanced.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistical Regression for Reference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>Coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>0.022100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>job</th>\n",
       "      <td>0.001677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>marital</th>\n",
       "      <td>0.085864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>education</th>\n",
       "      <td>0.019787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>default</th>\n",
       "      <td>0.246737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>housing</th>\n",
       "      <td>-0.004183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>loan</th>\n",
       "      <td>0.025915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>contact</th>\n",
       "      <td>-0.817695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>month</th>\n",
       "      <td>-0.113911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>campaign</th>\n",
       "      <td>-0.075320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pdays</th>\n",
       "      <td>-0.612735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous</th>\n",
       "      <td>-0.272723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>emp.var.rate</th>\n",
       "      <td>-1.657299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cons.price.idx</th>\n",
       "      <td>0.548991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cons.conf.idx</th>\n",
       "      <td>0.038687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>euribor3m</th>\n",
       "      <td>1.432062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nr.employed</th>\n",
       "      <td>-1.056283</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Coefficient\n",
       "age                0.022100\n",
       "job                0.001677\n",
       "marital            0.085864\n",
       "education          0.019787\n",
       "default            0.246737\n",
       "housing           -0.004183\n",
       "loan               0.025915\n",
       "contact           -0.817695\n",
       "month             -0.113911\n",
       "campaign          -0.075320\n",
       "pdays             -0.612735\n",
       "previous          -0.272723\n",
       "emp.var.rate      -1.657299\n",
       "cons.price.idx     0.548991\n",
       "cons.conf.idx      0.038687\n",
       "euribor3m          1.432062\n",
       "nr.employed       -1.056283"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "\n",
    "# Prepare features and target\n",
    "X_balanced = data_balanced.drop('subscribed', axis=1)\n",
    "y_balanced = data_balanced['subscribed']\n",
    "\n",
    "# Splitting the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_balanced, y_balanced, test_size=0.2, random_state=42)\n",
    "\n",
    "# Initialize Logistic Regression\n",
    "log_reg = LogisticRegression(max_iter=1000)\n",
    "\n",
    "# Fit the model\n",
    "log_reg.fit(X_train, y_train)\n",
    "\n",
    "# Predictions\n",
    "y_pred = log_reg.predict(X_test)\n",
    "\n",
    "# Evaluating the model\n",
    "report = classification_report(y_test, y_pred)\n",
    "conf_matrix = confusion_matrix(y_test, y_pred)\n",
    "\n",
    "# Coefficients\n",
    "coefficients = pd.DataFrame(log_reg.coef_.T, index=X_balanced.columns, columns=['Coefficient'])\n",
    "\n",
    "coefficients\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7344049247606019\n",
      "Precision: 0.7880129405755151\n",
      "Recall: 0.6369391687310763\n",
      "F1 Score: 0.704467615495852\n",
      "Confusion Matrix:\n",
      " [[6109 1245]\n",
      " [2638 4628]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix\n",
    "\n",
    "# Make predictions on the test set\n",
    "y_pred = log_reg.predict(X_test)\n",
    "\n",
    "# Calculate metrics\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "precision = precision_score(y_test, y_pred)\n",
    "recall = recall_score(y_test, y_pred)\n",
    "f1 = f1_score(y_test, y_pred)\n",
    "conf_matrix = confusion_matrix(y_test, y_pred)\n",
    "\n",
    "# Print the metrics\n",
    "print(\"Accuracy:\", accuracy)\n",
    "print(\"Precision:\", precision)\n",
    "print(\"Recall:\", recall)\n",
    "print(\"F1 Score:\", f1)\n",
    "print(\"Confusion Matrix:\\n\", conf_matrix)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Causal Inference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n"
     ]
    }
   ],
   "source": [
    "import econml\n",
    "from econml.dml import DML, LinearDML, SparseLinearDML, CausalForestDML\n",
    "from econml.orf import DMLOrthoForest\n",
    "import numpy as np\n",
    "from itertools import product\n",
    "from sklearn.linear_model import (Lasso, LassoCV, LogisticRegression,\n",
    "                                  LogisticRegressionCV,LinearRegression,\n",
    "                                  MultiTaskElasticNet,MultiTaskElasticNetCV)\n",
    "from sklearn.ensemble import RandomForestRegressor,RandomForestClassifier\n",
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "from sklearn.model_selection import train_test_split\n",
    "from xgboost import XGBRegressor\n",
    "from econml.sklearn_extensions.linear_model import WeightedLassoCVWrapper, WeightedLasso, WeightedLassoCV\n",
    "from sklearn.linear_model import LassoCV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Effect of Campaign"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "W = data_balanced[['age', 'job', 'marital', 'education', 'default', 'housing', 'loan', 'contact', 'cons.price.idx', 'cons.conf.idx', 'euribor3m']]\n",
    "T = dataset1[['campaign']]\n",
    "Y = data_balanced['subscribed']\n",
    "X = data_balanced[['age', 'job', 'marital', 'education', 'default', 'housing', 'loan', 'contact', 'month', 'cons.price.idx', 'cons.conf.idx', 'euribor3m']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# First split: Separating out the test set\n",
    "Y_train_val, Y_test, T_train_val, T_test, X_train_val, X_test, W_train_val, W_test = train_test_split(\n",
    "    Y, T, X, W, test_size=0.2, random_state=123)\n",
    "\n",
    "# Second split: Splitting the remaining data into training and validation sets\n",
    "Y_train, Y_val, T_train, T_val, X_train, X_val, W_train, W_val = train_test_split(\n",
    "    Y_train_val, T_train_val, X_train_val, W_train_val, test_size=0.25, random_state=123) # 0.25 * 0.8 = 0.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from econml.dml import CausalForestDML\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import shap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Casual Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean effect of treatment: -0.009241466081515031\n",
      "Effect Interval (95% confidence): [-0.026140881004831827, 0.0005314116355838516]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/05/03 00:43:48 WARNING mlflow.sklearn: Model was missing function: predict. Not logging python_function flavor!\n",
      "Registered model 'causal_forest_model' already exists. Creating a new version of this model...\n",
      "2024/05/03 00:43:57 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: causal_forest_model, version 3\n",
      "Created version '3' of model 'causal_forest_model'.\n"
     ]
    }
   ],
   "source": [
    "import mlflow\n",
    "from mlflow.tracking import MlflowClient\n",
    "import numpy as np\n",
    "from econml.dml import CausalForestDML\n",
    "from xgboost import XGBRegressor\n",
    "\n",
    "mlflow.set_tracking_uri(\"http://127.0.0.1:5000\")\n",
    "experiment_name = \"Campaign_Causal_Forest_Experiment\"\n",
    "client = MlflowClient()\n",
    "experiment = client.get_experiment_by_name(experiment_name)\n",
    "\n",
    "if experiment is None:\n",
    "    experiment_id = mlflow.create_experiment(experiment_name)\n",
    "else:\n",
    "    experiment_id = experiment.experiment_id\n",
    "\n",
    "with mlflow.start_run(experiment_id=experiment_id):\n",
    "    mlflow.log_param(\"model_type\", \"CausalForestDML\")\n",
    "    mlflow.log_param(\"model_y\", \"XGBRegressor\")\n",
    "    mlflow.log_param(\"model_y_seed\", 42)\n",
    "    mlflow.log_param(\"model_t\", \"XGBRegressor\")\n",
    "    mlflow.log_param(\"model_t_seed\", 42)\n",
    "    mlflow.log_param(\"criterion\", \"mse\")\n",
    "    mlflow.log_param(\"n_estimators\", 1000)\n",
    "    mlflow.log_param(\"min_impurity_decrease\", 0.0001)\n",
    "    mlflow.log_param(\"random_state\", 123)\n",
    "    \n",
    "    est = CausalForestDML(\n",
    "        model_y=XGBRegressor(objective='reg:squarederror', seed=42),\n",
    "        model_t=XGBRegressor(objective='reg:squarederror', seed=42),\n",
    "        criterion='mse', n_estimators=1000,\n",
    "        min_impurity_decrease=0.0001,\n",
    "        random_state=123\n",
    "    )\n",
    "    est.fit(Y_train, T_train, X=X_train, W=W_train)\n",
    "    \n",
    "    te_pred = est.effect(X_test)\n",
    "    \n",
    "    lower_bound = np.percentile(te_pred, 2.5)\n",
    "    upper_bound = np.percentile(te_pred, 97.5)\n",
    "    \n",
    "    avg_effect = est.ate(X)\n",
    "\n",
    "\n",
    "    mlflow.log_metric(\"average_treatment_effect\", avg_effect)\n",
    "    mlflow.log_metric(\"lower_bound_95_CI\", lower_bound)\n",
    "    mlflow.log_metric(\"upper_bound_95_CI\", upper_bound)\n",
    "    \n",
    "    print(f\"Mean effect of treatment: {avg_effect}\")\n",
    "    print(f\"Effect Interval (95% confidence): [{lower_bound}, {upper_bound}]\")\n",
    "\n",
    "    mlflow.sklearn.log_model(est, \"causal_forest_model\", registered_model_name=\"causal_forest_model\")\n",
    "    mlflow.end_run()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linear DML"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/05/03 00:43:57 INFO mlflow.tracking.fluent: Experiment with name 'Campaign_Linear_DML_Experiment' does not exist. Creating a new experiment.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/05/03 00:43:58 WARNING mlflow.sklearn: Model was missing function: predict. Not logging python_function flavor!\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean effect of treatment: -0.01135688211468841\n",
      "Effect Interval (95% confidence): [-0.03639147084816514, 0.008897183497642394]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Registered model 'linear_dml_model' already exists. Creating a new version of this model...\n",
      "2024/05/03 00:44:06 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: linear_dml_model, version 2\n",
      "Created version '2' of model 'linear_dml_model'.\n"
     ]
    }
   ],
   "source": [
    "import mlflow\n",
    "from econml.dml import LinearDML\n",
    "from xgboost import XGBRegressor\n",
    "import numpy as np\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "mlflow.set_tracking_uri(\"http://127.0.0.1:5000\")\n",
    "experiment_name = \"Campaign_Linear_DML_Experiment\"\n",
    "mlflow.set_experiment(experiment_name)\n",
    "\n",
    "with mlflow.start_run():\n",
    "    mlflow.log_params({\n",
    "        \"model_type\": \"LinearDML\",\n",
    "        \"model_y\": \"XGBRegressor\",\n",
    "        \"model_y_seed\": 42,\n",
    "        \"model_t\": \"XGBRegressor\",\n",
    "        \"model_t_seed\": 42,\n",
    "        \"random_state\": 123\n",
    "    })\n",
    "\n",
    "    est2 = LinearDML(\n",
    "        model_y=XGBRegressor(objective='reg:squarederror', seed=42),\n",
    "        model_t=XGBRegressor(objective='reg:squarederror', seed=42),\n",
    "        random_state=123\n",
    "    )\n",
    "    \n",
    "\n",
    "    est2.fit(Y_train, T_train, X=X_train, W=W_train)\n",
    "    \n",
    "    te_pred2 = est2.effect(X_test)\n",
    "\n",
    "    lower_bound2 = np.percentile(te_pred2, 2.5)\n",
    "    upper_bound2 = np.percentile(te_pred2, 97.5)\n",
    "    \n",
    "    avg_effect2 = est2.ate(X)\n",
    "    \n",
    "    mlflow.log_metrics({\n",
    "        \"average_treatment_effect\": avg_effect2,\n",
    "        \"lower_bound_95_CI\": lower_bound2,\n",
    "        \"upper_bound_95_CI\": upper_bound2\n",
    "    })\n",
    "    \n",
    "    print(f\"Mean effect of treatment: {avg_effect2}\")\n",
    "    print(f\"Effect Interval (95% confidence): [{lower_bound2}, {upper_bound2}]\")\n",
    "\n",
    "    mlflow.sklearn.log_model(est2, \"linear_dml_model\", registered_model_name=\"linear_dml_model\")\n",
    "    mlflow.end_run()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "shap_values = est2.shap_values(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x630 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# convert dict_keys object to a list to make it subscriptable\n",
    "output_name = list(shap_values.keys())[0] \n",
    "treatment_name = list(shap_values[output_name].keys())[0] \n",
    "\n",
    "specific_shap_values = shap_values[output_name][treatment_name]\n",
    "\n",
    "\n",
    "# visualize the SHAP values\n",
    "shap.summary_plot(specific_shap_values, X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Effect of Euribor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "W = data_balanced[['age', 'job', 'marital', 'education', 'default', 'housing', 'loan', 'contact', 'cons.price.idx', 'cons.conf.idx','campaign']]\n",
    "T = dataset1[['euribor3m']]\n",
    "Y = data_balanced['subscribed']\n",
    "X = data_balanced[['age', 'job', 'marital', 'education', 'default', 'housing', 'loan', 'contact', 'month', 'cons.price.idx', 'cons.conf.idx','campaign']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# First split: Separating out the test set\n",
    "Y_train_val, Y_test, T_train_val, T_test, X_train_val, X_test, W_train_val, W_test = train_test_split(\n",
    "    Y, T, X, W, test_size=0.2, random_state=123)\n",
    "\n",
    "# Second split: Splitting the remaining data into training and validation sets\n",
    "Y_train, Y_val, T_train, T_val, X_train, X_val, W_train, W_val = train_test_split(\n",
    "    Y_train_val, T_train_val, X_train_val, W_train_val, test_size=0.25, random_state=123) # 0.25 * 0.8 = 0.2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Casual Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean effect of treatment:-0.10922887674412755\n",
      "Effect Interval (95% confidence): [-0.3703344134277583, -0.030699659134896196]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/05/03 00:50:19 WARNING mlflow.sklearn: Model was missing function: predict. Not logging python_function flavor!\n",
      "Registered model 'CausalForestDML' already exists. Creating a new version of this model...\n",
      "2024/05/03 00:50:27 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: CausalForestDML, version 3\n",
      "Created version '3' of model 'CausalForestDML'.\n"
     ]
    }
   ],
   "source": [
    "import mlflow\n",
    "from econml.dml import LinearDML\n",
    "from xgboost import XGBRegressor\n",
    "import numpy as np\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "mlflow.set_tracking_uri(\"http://127.0.0.1:5000\")\n",
    "experiment_name = \"Euribor_Causal_Forest_Experiment\"\n",
    "mlflow.set_experiment(experiment_name)\n",
    "\n",
    "with mlflow.start_run():\n",
    "    mlflow.log_params({\n",
    "        \"model_type\": \"CausalForestDML\",\n",
    "        \"model_y\": \"XGBRegressor\",\n",
    "        \"model_y_seed\": 42,\n",
    "        \"model_t\": \"XGBRegressor\",\n",
    "        \"model_t_seed\": 42,\n",
    "        \"random_state\": 123,\n",
    "        \"min_impurity_decrease\": 0.0005,\n",
    "        \"n_estimators\": 1000,\n",
    "        \"criterion\": \"mse\"   \n",
    "    })\n",
    "    \n",
    "\n",
    "\n",
    "    est = CausalForestDML(model_y=XGBRegressor(objective='reg:squarederror', seed=42),\n",
    "                        model_t=XGBRegressor(objective='reg:squarederror', seed=42),\n",
    "                        criterion='mse', n_estimators=1000,\n",
    "                        min_impurity_decrease=0.0005,\n",
    "                        random_state=123)\n",
    "    est.fit(Y_train, T_train, X=X_train, W=W_train)\n",
    "    te_pred = est.effect(X_test)\n",
    "    \n",
    "\n",
    "    # Calculate the confidence interval using percentiles\n",
    "    lower_bound = np.percentile(te_pred, 2.5)\n",
    "    upper_bound = np.percentile(te_pred, 97.5)\n",
    "    avg_effect = est.ate(X)\n",
    "\n",
    "    mlflow.log_metrics({\n",
    "        \"average_treatment_effect\": avg_effect,\n",
    "        \"lower_bound_95_CI\": lower_bound,\n",
    "        \"upper_bound_95_CI\": upper_bound\n",
    "    })\n",
    "    \n",
    "    print(f\"Mean effect of treatment:{avg_effect}\")\n",
    "    print(\"Effect Interval (95% confidence): [{}, {}]\".format(lower_bound, upper_bound))\n",
    "\n",
    "\n",
    "    mlflow.sklearn.log_model(est, \"CausalForestDML\", registered_model_name=\"CausalForestDML\")\n",
    "    mlflow.end_run()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linear DML"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/05/03 00:52:10 INFO mlflow.tracking.fluent: Experiment with name 'Eurbior_Linear_DML_Experiment' does not exist. Creating a new experiment.\n",
      "2024/05/03 00:52:10 WARNING mlflow.sklearn: Model was missing function: predict. Not logging python_function flavor!\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean effect of treatment:-0.11105254221583588\n",
      "Effect Interval (95% confidence): [-0.18311508062714998, -0.015680850961249937]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Registered model 'linear_dml_model' already exists. Creating a new version of this model...\n",
      "2024/05/03 00:52:18 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: linear_dml_model, version 3\n",
      "Created version '3' of model 'linear_dml_model'.\n"
     ]
    }
   ],
   "source": [
    "import mlflow\n",
    "from econml.dml import LinearDML\n",
    "from xgboost import XGBRegressor\n",
    "import numpy as np\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "mlflow.set_tracking_uri(\"http://127.0.0.1:5000\")\n",
    "experiment_name = \"Eurbior_Linear_DML_Experiment\"\n",
    "mlflow.set_experiment(experiment_name)\n",
    "\n",
    "with mlflow.start_run():\n",
    "    mlflow.log_params({\n",
    "        \"model_type\": \"LinearDML\",\n",
    "        \"model_y\": \"Lasso\",\n",
    "        \"model_t\": \"Lasso\",\n",
    "        \"random_state\": 123\n",
    "    })\n",
    "    \n",
    "    est2 = LinearDML(model_y=Lasso(),\n",
    "                    model_t=Lasso(),\n",
    "                    random_state=123)\n",
    "    est2.fit(Y_train, T_train, X=X_train, W=W_train)\n",
    "    te_pred2 = est2.effect(X_test)\n",
    "\n",
    "    est2.fit(Y_train, T_train, X=X_train, W=W_train)\n",
    "    \n",
    "    te_pred2 = est2.effect(X_test)\n",
    "\n",
    "    # Calculate the confidence interval using percentiles\n",
    "    lower_bound2 = np.percentile(te_pred2, 2.5)\n",
    "    upper_bound2 = np.percentile(te_pred2, 97.5)\n",
    "    avg_effect2 = est2.ate(X)\n",
    "    # Print effect\n",
    "    print(f\"Mean effect of treatment:{avg_effect2}\")\n",
    "    print(\"Effect Interval (95% confidence): [{}, {}]\".format(lower_bound2, upper_bound2))\n",
    "    \n",
    "    mlflow.log_metrics({\n",
    "        \"average_treatment_effect\": avg_effect2,\n",
    "        \"lower_bound_95_CI\": lower_bound2,\n",
    "        \"upper_bound_95_CI\": upper_bound2\n",
    "    })\n",
    "\n",
    "    mlflow.sklearn.log_model(est2, \"linear_dml_model\", registered_model_name=\"linear_dml_model\")\n",
    "    mlflow.end_run()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "shap_values = est2.shap_values(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x630 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# convert dict_keys object to a list to make it subscriptable\n",
    "output_name = list(shap_values.keys())[0] \n",
    "treatment_name = list(shap_values[output_name].keys())[0] \n",
    "\n",
    "specific_shap_values = shap_values[output_name][treatment_name]\n",
    "\n",
    "\n",
    "# visualize the SHAP values\n",
    "shap.summary_plot(specific_shap_values, X)"
   ]
  }
 ],
 "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": "3.12.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}