PD Model with scorecard.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading the Data and Selecting the Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_inputs_train = pd.read_csv('loan_data_inputs_train.csv', index_col = 0)\n",
    "loan_data_targets_train = pd.read_csv('loan_data_targets_train.csv', index_col = 0)\n",
    "loan_data_inputs_test = pd.read_csv('loan_data_inputs_test.csv', index_col = 0)\n",
    "loan_data_targets_test = pd.read_csv('loan_data_targets_test.csv', index_col = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Explore Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_inputs_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_inputs_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_inputs_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Selecting the Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "inputs_train_with_ref_cat = loan_data_inputs_train.reindex(columns=['grade:A','grade:B',\n",
    "'grade:C',\n",
    "'grade:D',\n",
    "'grade:E',\n",
    "'grade:F',\n",
    "'grade:G',\n",
    "'home_ownership:RENT_OTHER_NONE_ANY',\n",
    "'home_ownership:OWN',\n",
    "'home_ownership:MORTGAGE',\n",
    "'addr_state:ND_NE_IA_NV_FL_HI_AL',\n",
    "'addr_state:NM_VA',\n",
    "'addr_state:NY',\n",
    "'addr_state:OK_TN_MO_LA_MD_NC',\n",
    "'addr_state:CA',\n",
    "'addr_state:UT_KY_AZ_NJ',\n",
    "'addr_state:AR_MI_PA_OH_MN',\n",
    "'addr_state:RI_MA_DE_SD_IN',\n",
    "'addr_state:GA_WA_OR',\n",
    "'addr_state:WI_MT',\n",
    "'addr_state:TX',\n",
    "'addr_state:IL_CT',\n",
    "'addr_state:KS_SC_CO_VT_AK_MS',\n",
    "'addr_state:WV_NH_WY_DC_ME_ID',\n",
    "'verification_status:Not Verified',\n",
    "'verification_status:Source Verified',\n",
    "'verification_status:Verified',\n",
    "'purpose:educ__sm_b__wedd__ren_en__mov__house',\n",
    "'purpose:credit_card',\n",
    "'purpose:debt_consolidation',\n",
    "'purpose:oth__med__vacation',\n",
    "'purpose:major_purch__car__home_impr',\n",
    "'initial_list_status:f',\n",
    "'initial_list_status:w',\n",
    "'term:36',\n",
    "'term:60',\n",
    "'emp_length:0',\n",
    "'emp_length:1',\n",
    "'emp_length:2-4',\n",
    "'emp_length:5-6',\n",
    "'emp_length:7-9',\n",
    "'emp_length:10',\n",
    "'mths_since_issue_d:<38',\n",
    "'mths_since_issue_d:38-39',\n",
    "'mths_since_issue_d:40-41',\n",
    "'mths_since_issue_d:42-48',\n",
    "'mths_since_issue_d:49-52',\n",
    "'mths_since_issue_d:53-64',\n",
    "'mths_since_issue_d:65-84',\n",
    "'mths_since_issue_d:>84',\n",
    "'int_rate:<9.548',\n",
    "'int_rate:9.548-12.025',\n",
    "'int_rate:12.025-15.74',\n",
    "'int_rate:15.74-20.281',\n",
    "'int_rate:>20.281',\n",
    "'mths_since_earliest_cr_line:<140',\n",
    "'mths_since_earliest_cr_line:141-164',\n",
    "'mths_since_earliest_cr_line:165-247',\n",
    "'mths_since_earliest_cr_line:248-270',\n",
    "'mths_since_earliest_cr_line:271-352',\n",
    "'mths_since_earliest_cr_line:>352',\n",
    "'delinq_2yrs:0',\n",
    "'delinq_2yrs:1-3',\n",
    "'delinq_2yrs:>=4',\n",
    "'inq_last_6mths:0',\n",
    "'inq_last_6mths:1-2',\n",
    "'inq_last_6mths:3-6',\n",
    "'inq_last_6mths:>6',\n",
    "'open_acc:0',\n",
    "'open_acc:1-3',\n",
    "'open_acc:4-12',\n",
    "'open_acc:13-17',\n",
    "'open_acc:18-22',\n",
    "'open_acc:23-25',\n",
    "'open_acc:26-30',\n",
    "'open_acc:>=31',\n",
    "'pub_rec:0-2',\n",
    "'pub_rec:3-4',\n",
    "'pub_rec:>=5',\n",
    "'total_acc:<=27',\n",
    "'total_acc:28-51',\n",
    "'total_acc:>=52',\n",
    "'acc_now_delinq:0',\n",
    "'acc_now_delinq:>=1',\n",
    "'total_rev_hi_lim:<=5K',\n",
    "'total_rev_hi_lim:5K-10K',\n",
    "'total_rev_hi_lim:10K-20K',\n",
    "'total_rev_hi_lim:20K-30K',\n",
    "'total_rev_hi_lim:30K-40K',\n",
    "'total_rev_hi_lim:40K-55K',\n",
    "'total_rev_hi_lim:55K-95K',\n",
    "'total_rev_hi_lim:>95K',\n",
    "'annual_inc:<20K',\n",
    "'annual_inc:20K-30K',\n",
    "'annual_inc:30K-40K',\n",
    "'annual_inc:40K-50K',\n",
    "'annual_inc:50K-60K',\n",
    "'annual_inc:60K-70K',\n",
    "'annual_inc:70K-80K',\n",
    "'annual_inc:80K-90K',\n",
    "'annual_inc:90K-100K',\n",
    "'annual_inc:100K-120K',\n",
    "'annual_inc:120K-140K',\n",
    "'annual_inc:>140K',\n",
    "'dti:<=1.4',\n",
    "'dti:1.4-3.5',\n",
    "'dti:3.5-7.7',\n",
    "'dti:7.7-10.5',\n",
    "'dti:10.5-16.1',\n",
    "'dti:16.1-20.3',\n",
    "'dti:20.3-21.7',\n",
    "'dti:21.7-22.4',\n",
    "'dti:22.4-35',\n",
    "'dti:>35',\n",
    "'mths_since_last_delinq:Missing',\n",
    "'mths_since_last_delinq:0-3',\n",
    "'mths_since_last_delinq:4-30',\n",
    "'mths_since_last_delinq:31-56',\n",
    "'mths_since_last_delinq:>=57',\n",
    "'mths_since_last_record:Missing',\n",
    "'mths_since_last_record:0-2',\n",
    "'mths_since_last_record:3-20',\n",
    "'mths_since_last_record:21-31',\n",
    "'mths_since_last_record:32-80',\n",
    "'mths_since_last_record:81-86',\n",
    "'mths_since_last_record:>86',\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ref_categories = ['grade:G',\n",
    "'home_ownership:RENT_OTHER_NONE_ANY',\n",
    "'addr_state:ND_NE_IA_NV_FL_HI_AL',\n",
    "'verification_status:Verified',\n",
    "'purpose:educ__sm_b__wedd__ren_en__mov__house',\n",
    "'initial_list_status:f',\n",
    "'term:60',\n",
    "'emp_length:0',\n",
    "'mths_since_issue_d:>84',\n",
    "'int_rate:>20.281',\n",
    "'mths_since_earliest_cr_line:<140',\n",
    "'delinq_2yrs:>=4',\n",
    "'inq_last_6mths:>6',\n",
    "'open_acc:0',\n",
    "'pub_rec:0-2',\n",
    "'total_acc:<=27',\n",
    "'acc_now_delinq:0',\n",
    "'total_rev_hi_lim:<=5K',\n",
    "'annual_inc:<20K',\n",
    "'dti:>35',\n",
    "'mths_since_last_delinq:0-3',\n",
    "'mths_since_last_record:0-2']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "inputs_train = inputs_train_with_ref_cat.drop(ref_categories, axis = 1)\n",
    "# From the dataframe with input variables, we drop the variables with variable names in the list with reference categories. \n",
    "inputs_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PD Model Estimation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_train.fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_train.fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import metrics\n",
    "import sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = LogisticRegression()\n",
    "# We create an instance of an object from the 'LogisticRegression' class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.options.display.max_rows = None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "reg.fit(inputs_train, loan_data_targets_train.values.ravel())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "reg.intercept_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "reg.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_name = inputs_train.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_table = pd.DataFrame(columns = ['Feature name'], data = feature_name)\n",
    "summary_table['Coefficients'] = np.transpose(reg.coef_)\n",
    "summary_table.index = summary_table.index + 1\n",
    "summary_table.loc[0] = ['Intercept', reg.intercept_[0]]\n",
    "summary_table = summary_table.sort_index()\n",
    "summary_table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build a Logistic Regression Model with P-Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn.ensemble\n",
    "from sklearn import linear_model\n",
    "import scipy.stats as stat\n",
    "\n",
    "class LogisticRegression_with_p_values:\n",
    "    \n",
    "    def __init__(self,*args,**kwargs):#,**kwargs):\n",
    "        self.model = linear_model.LogisticRegression(*args,**kwargs)#,**args)\n",
    "\n",
    "    def fit(self,X,y):\n",
    "        self.model.fit(X,y)\n",
    "\n",
    "        denom = (2.0 * (1.0 + np.cosh(self.model.decision_function(X))))\n",
    "        denom = np.tile(denom,(X.shape[1],1)).T\n",
    "        F_ij = np.dot((X / denom).T,X) ## Fisher Information Matrix\n",
    "        Cramer_Rao = np.linalg.pinv(F_ij) ## Inverse Information Matrix\n",
    "        sigma_estimates = np.sqrt(np.diagonal(Cramer_Rao))\n",
    "        z_scores = self.model.coef_[0] / sigma_estimates # z-score for eaach model coefficient\n",
    "        p_values = [stat.norm.sf(abs(x)) * 2 for x in z_scores] ### two tailed test for p-values\n",
    "        \n",
    "        self.coef_ = self.model.coef_\n",
    "        self.intercept_ = self.model.intercept_\n",
    "        self.p_values = p_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import linear_model\n",
    "import scipy.stats as stat\n",
    "import numpy\n",
    "class LogisticRegression_with_p_values:\n",
    "    \n",
    "    def __init__(self,*args,**kwargs):\n",
    "        self.model = linear_model.LogisticRegression(*args,**kwargs)\n",
    "\n",
    "    def fit(self,X,y):\n",
    "        self.model.fit(X,y)\n",
    "        denom = (2.0 * (1.0 + np.cosh(self.model.decision_function(X))))\n",
    "        denom = np.tile(denom,(X.shape[1],1)).T\n",
    "        F_ij = np.dot((X / denom).T,X)\n",
    "        Cramer_Rao = np.linalg.inv(F_ij)\n",
    "        sigma_estimates = np.sqrt(np.diagonal(Cramer_Rao))\n",
    "        z_scores = self.model.coef_[0] / sigma_estimates\n",
    "        p_values = [stat.norm.sf(abs(x)) * 2 for x in z_scores]\n",
    "        self.coef_ = self.model.coef_\n",
    "        self.intercept_ = self.model.intercept_\n",
    "        self.p_values = p_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = LogisticRegression_with_p_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_train=pd.DataFrame(loan_data_targets_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "reg.fit(inputs_train, loan_data_targets_train.values.ravel())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "summary_table = pd.DataFrame(columns = ['Feature name'], data = feature_name)\n",
    "summary_table['Coefficients'] = np.transpose(reg.coef_)\n",
    "summary_table.index = summary_table.index + 1\n",
    "summary_table.loc[0] = ['Intercept', reg.intercept_[0]]\n",
    "summary_table = summary_table.sort_index()\n",
    "summary_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "p_values = reg.p_values\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "p_values = np.append(np.nan, np.array(p_values))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_table['p_values'] = p_values\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Variables\n",
    "inputs_train_with_ref_cat = loan_data_inputs_train.reindex(columns=['grade:A','grade:B',\n",
    "'grade:C',\n",
    "'grade:D',\n",
    "'grade:E',\n",
    "'grade:F',\n",
    "'grade:G',\n",
    "'home_ownership:RENT_OTHER_NONE_ANY',\n",
    "'home_ownership:OWN',\n",
    "'home_ownership:MORTGAGE',\n",
    "'addr_state:ND_NE_IA_NV_FL_HI_AL',\n",
    "'addr_state:NM_VA',\n",
    "'addr_state:NY',\n",
    "'addr_state:OK_TN_MO_LA_MD_NC',\n",
    "'addr_state:CA',\n",
    "'addr_state:UT_KY_AZ_NJ',\n",
    "'addr_state:AR_MI_PA_OH_MN',\n",
    "'addr_state:RI_MA_DE_SD_IN',\n",
    "'addr_state:GA_WA_OR',\n",
    "'addr_state:WI_MT',\n",
    "'addr_state:TX',\n",
    "'addr_state:IL_CT',\n",
    "'addr_state:KS_SC_CO_VT_AK_MS',\n",
    "'addr_state:WV_NH_WY_DC_ME_ID',\n",
    "'verification_status:Not Verified',\n",
    "'verification_status:Source Verified',\n",
    "'verification_status:Verified',\n",
    "'purpose:educ__sm_b__wedd__ren_en__mov__house',\n",
    "'purpose:credit_card',\n",
    "'purpose:debt_consolidation',\n",
    "'purpose:oth__med__vacation',\n",
    "'purpose:major_purch__car__home_impr',\n",
    "'initial_list_status:f',\n",
    "'initial_list_status:w',\n",
    "'term:36',\n",
    "'term:60',\n",
    "'emp_length:0',\n",
    "'emp_length:1',\n",
    "'emp_length:2-4',\n",
    "'emp_length:5-6',\n",
    "'emp_length:7-9',\n",
    "'emp_length:10',\n",
    "'mths_since_issue_d:<38',\n",
    "'mths_since_issue_d:38-39',\n",
    "'mths_since_issue_d:40-41',\n",
    "'mths_since_issue_d:42-48',\n",
    "'mths_since_issue_d:49-52',\n",
    "'mths_since_issue_d:53-64',\n",
    "'mths_since_issue_d:65-84',\n",
    "'mths_since_issue_d:>84',\n",
    "'int_rate:<9.548',\n",
    "'int_rate:9.548-12.025',\n",
    "'int_rate:12.025-15.74',\n",
    "'int_rate:15.74-20.281',\n",
    "'int_rate:>20.281',\n",
    "'mths_since_earliest_cr_line:<140',\n",
    "'mths_since_earliest_cr_line:141-164',\n",
    "'mths_since_earliest_cr_line:165-247',\n",
    "'mths_since_earliest_cr_line:248-270',\n",
    "'mths_since_earliest_cr_line:271-352',\n",
    "'mths_since_earliest_cr_line:>352',\n",
    "'inq_last_6mths:0',\n",
    "'inq_last_6mths:1-2',\n",
    "'inq_last_6mths:3-6',\n",
    "'inq_last_6mths:>6',\n",
    "'acc_now_delinq:0',\n",
    "'acc_now_delinq:>=1',\n",
    "'annual_inc:<20K',\n",
    "'annual_inc:20K-30K',\n",
    "'annual_inc:30K-40K',\n",
    "'annual_inc:40K-50K',\n",
    "'annual_inc:50K-60K',\n",
    "'annual_inc:60K-70K',\n",
    "'annual_inc:70K-80K',\n",
    "'annual_inc:80K-90K',\n",
    "'annual_inc:90K-100K',\n",
    "'annual_inc:100K-120K',\n",
    "'annual_inc:120K-140K',\n",
    "'annual_inc:>140K',\n",
    "'dti:<=1.4',\n",
    "'dti:1.4-3.5',\n",
    "'dti:3.5-7.7',\n",
    "'dti:7.7-10.5',\n",
    "'dti:10.5-16.1',\n",
    "'dti:16.1-20.3',\n",
    "'dti:20.3-21.7',\n",
    "'dti:21.7-22.4',\n",
    "'dti:22.4-35',\n",
    "'dti:>35',\n",
    "'mths_since_last_delinq:Missing',\n",
    "'mths_since_last_delinq:0-3',\n",
    "'mths_since_last_delinq:4-30',\n",
    "'mths_since_last_delinq:31-56',\n",
    "'mths_since_last_delinq:>=57',\n",
    "'mths_since_last_record:Missing',\n",
    "'mths_since_last_record:0-2',\n",
    "'mths_since_last_record:3-20',\n",
    "'mths_since_last_record:21-31',\n",
    "'mths_since_last_record:32-80',\n",
    "'mths_since_last_record:81-86',\n",
    "'mths_since_last_record:>86',\n",
    "])\n",
    "\n",
    "ref_categories = ['grade:G',\n",
    "'home_ownership:RENT_OTHER_NONE_ANY',\n",
    "'addr_state:ND_NE_IA_NV_FL_HI_AL',\n",
    "'verification_status:Verified',\n",
    "'purpose:educ__sm_b__wedd__ren_en__mov__house',\n",
    "'initial_list_status:f',\n",
    "'term:60',\n",
    "'emp_length:0',\n",
    "'mths_since_issue_d:>84',\n",
    "'int_rate:>20.281',\n",
    "'mths_since_earliest_cr_line:<140',\n",
    "'inq_last_6mths:>6',\n",
    "'acc_now_delinq:0',\n",
    "'annual_inc:<20K',\n",
    "'dti:>35',\n",
    "'mths_since_last_delinq:0-3',\n",
    "'mths_since_last_record:0-2']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "inputs_train = inputs_train_with_ref_cat.drop(ref_categories, axis = 1)\n",
    "inputs_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "reg2 = LogisticRegression_with_p_values()\n",
    "reg2.fit(inputs_train, loan_data_targets_train.values.ravel())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_name = inputs_train.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "summary_table = pd.DataFrame(columns = ['Feature name'], data = feature_name)\n",
    "summary_table['Coefficients'] = np.transpose(reg2.coef_)\n",
    "summary_table.index = summary_table.index + 1\n",
    "summary_table.loc[0] = ['Intercept', reg2.intercept_[0]]\n",
    "summary_table = summary_table.sort_index()\n",
    "summary_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "p_values = reg2.p_values\n",
    "p_values = np.append(np.nan,np.array(p_values))\n",
    "summary_table['p_values'] = p_values\n",
    "summary_table\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pickle.dump(reg2, open('pd_model.sav', 'wb'))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PD Model Validation (Test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Out-of-sample validation (test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Here, from the dataframe with inputs for testing, we keep the same variables that we used in our final PD model.\n",
    "inputs_test_with_ref_cat = loan_data_inputs_test.reindex(columns=['grade:A','grade:B',\n",
    "'grade:C',\n",
    "'grade:D',\n",
    "'grade:E',\n",
    "'grade:F',\n",
    "'grade:G',\n",
    "'home_ownership:RENT_OTHER_NONE_ANY',\n",
    "'home_ownership:OWN',\n",
    "'home_ownership:MORTGAGE',\n",
    "'addr_state:ND_NE_IA_NV_FL_HI_AL',\n",
    "'addr_state:NM_VA',\n",
    "'addr_state:NY',\n",
    "'addr_state:OK_TN_MO_LA_MD_NC',\n",
    "'addr_state:CA',\n",
    "'addr_state:UT_KY_AZ_NJ',\n",
    "'addr_state:AR_MI_PA_OH_MN',\n",
    "'addr_state:RI_MA_DE_SD_IN',\n",
    "'addr_state:GA_WA_OR',\n",
    "'addr_state:WI_MT',\n",
    "'addr_state:TX',\n",
    "'addr_state:IL_CT',\n",
    "'addr_state:KS_SC_CO_VT_AK_MS',\n",
    "'addr_state:WV_NH_WY_DC_ME_ID',\n",
    "'verification_status:Not Verified',\n",
    "'verification_status:Source Verified',\n",
    "'verification_status:Verified',\n",
    "'purpose:educ__sm_b__wedd__ren_en__mov__house',\n",
    "'purpose:credit_card',\n",
    "'purpose:debt_consolidation',\n",
    "'purpose:oth__med__vacation',\n",
    "'purpose:major_purch__car__home_impr',\n",
    "'initial_list_status:f',\n",
    "'initial_list_status:w',\n",
    "'term:36',\n",
    "'term:60',\n",
    "'emp_length:0',\n",
    "'emp_length:1',\n",
    "'emp_length:2-4',\n",
    "'emp_length:5-6',\n",
    "'emp_length:7-9',\n",
    "'emp_length:10',\n",
    "'mths_since_issue_d:<38',\n",
    "'mths_since_issue_d:38-39',\n",
    "'mths_since_issue_d:40-41',\n",
    "'mths_since_issue_d:42-48',\n",
    "'mths_since_issue_d:49-52',\n",
    "'mths_since_issue_d:53-64',\n",
    "'mths_since_issue_d:65-84',\n",
    "'mths_since_issue_d:>84',\n",
    "'int_rate:<9.548',\n",
    "'int_rate:9.548-12.025',\n",
    "'int_rate:12.025-15.74',\n",
    "'int_rate:15.74-20.281',\n",
    "'int_rate:>20.281',\n",
    "'mths_since_earliest_cr_line:<140',\n",
    "'mths_since_earliest_cr_line:141-164',\n",
    "'mths_since_earliest_cr_line:165-247',\n",
    "'mths_since_earliest_cr_line:248-270',\n",
    "'mths_since_earliest_cr_line:271-352',\n",
    "'mths_since_earliest_cr_line:>352',\n",
    "'inq_last_6mths:0',\n",
    "'inq_last_6mths:1-2',\n",
    "'inq_last_6mths:3-6',\n",
    "'inq_last_6mths:>6',\n",
    "'acc_now_delinq:0',\n",
    "'acc_now_delinq:>=1',\n",
    "'annual_inc:<20K',\n",
    "'annual_inc:20K-30K',\n",
    "'annual_inc:30K-40K',\n",
    "'annual_inc:40K-50K',\n",
    "'annual_inc:50K-60K',\n",
    "'annual_inc:60K-70K',\n",
    "'annual_inc:70K-80K',\n",
    "'annual_inc:80K-90K',\n",
    "'annual_inc:90K-100K',\n",
    "'annual_inc:100K-120K',\n",
    "'annual_inc:120K-140K',\n",
    "'annual_inc:>140K',\n",
    "'dti:<=1.4',\n",
    "'dti:1.4-3.5',\n",
    "'dti:3.5-7.7',\n",
    "'dti:7.7-10.5',\n",
    "'dti:10.5-16.1',\n",
    "'dti:16.1-20.3',\n",
    "'dti:20.3-21.7',\n",
    "'dti:21.7-22.4',\n",
    "'dti:22.4-35',\n",
    "'dti:>35',\n",
    "'mths_since_last_delinq:Missing',\n",
    "'mths_since_last_delinq:0-3',\n",
    "'mths_since_last_delinq:4-30',\n",
    "'mths_since_last_delinq:31-56',\n",
    "'mths_since_last_delinq:>=57',\n",
    "'mths_since_last_record:Missing',\n",
    "'mths_since_last_record:0-2',\n",
    "'mths_since_last_record:3-20',\n",
    "'mths_since_last_record:21-31',\n",
    "'mths_since_last_record:32-80',\n",
    "'mths_since_last_record:81-86',\n",
    "'mths_since_last_record:>86',\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "ref_categories = ['grade:G',\n",
    "'home_ownership:RENT_OTHER_NONE_ANY',\n",
    "'addr_state:ND_NE_IA_NV_FL_HI_AL',\n",
    "'verification_status:Verified',\n",
    "'purpose:educ__sm_b__wedd__ren_en__mov__house',\n",
    "'initial_list_status:f',\n",
    "'term:60',\n",
    "'emp_length:0',\n",
    "'mths_since_issue_d:>84',\n",
    "'int_rate:>20.281',\n",
    "'mths_since_earliest_cr_line:<140',\n",
    "'inq_last_6mths:>6',\n",
    "'acc_now_delinq:0',\n",
    "'annual_inc:<20K',\n",
    "'dti:>35',\n",
    "'mths_since_last_delinq:0-3',\n",
    "'mths_since_last_record:0-2']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test = inputs_test_with_ref_cat.drop(ref_categories, axis = 1)\n",
    "inputs_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test = reg2.model.predict(inputs_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba = reg2.model.predict_proba(inputs_test)\n",
    "# Calculates the predicted probability values for the dependent variable (targets)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba[:][:,1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba = y_hat_test_proba[: ][: , 1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_test_temp = loan_data_targets_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loan_data_targets_test_temp.reset_index(drop = True, inplace = True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs = pd.concat([loan_data_targets_test_temp, pd.DataFrame(y_hat_test_proba)], axis = 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.columns = ['loan_data_targets_test', 'y_hat_test_proba']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.index = loan_data_inputs_test.index\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Accuracy and Area under the Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tr = 0.9\n",
    "\n",
    "df_actual_predicted_probs['y_hat_test'] = np.where(df_actual_predicted_probs['y_hat_test_proba'] > tr, 1, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted']) / df_actual_predicted_probs.shape[0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted']) / df_actual_predicted_probs.shape[0]).iloc[0, 0] + (pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted']) / df_actual_predicted_probs.shape[0]).iloc[1, 1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "roc_curve(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test_proba'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fpr, tpr, thresholds = roc_curve(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test_proba'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "plt.plot(fpr, tpr)\n",
    "# We plot the false positive rate along the x-axis and the true positive rate along the y-axis,\n",
    "# thus plotting the ROC curve.\n",
    "plt.plot(fpr, fpr, linestyle = '--', color = 'k')\n",
    "# We plot a seconary diagonal line, with dashed line style and black color.\n",
    "plt.xlabel('False positive rate')\n",
    "# We name the x-axis \"False positive rate\".\n",
    "plt.ylabel('True positive rate')\n",
    "# We name the x-axis \"True positive rate\".\n",
    "plt.title('ROC curve')\n",
    "# We name the graph \"ROC curve\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "AUROC = roc_auc_score(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test_proba'])\n",
    "\n",
    "AUROC"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gini and Kolmogorov-Smirnov"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs = df_actual_predicted_probs.sort_values('y_hat_test_proba')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs = df_actual_predicted_probs.reset_index()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs['Cumulative N Population'] = df_actual_predicted_probs.index + 1\n",
    "# We calculate the cumulative number of all observations.\n",
    "# We use the new index for that. Since indexing in ython starts from 0, we add 1 to each index.\n",
    "df_actual_predicted_probs['Cumulative N Good'] = df_actual_predicted_probs['loan_data_targets_test'].cumsum()\n",
    "# We calculate cumulative number of 'good', which is the cumulative sum of the column with actual observations.\n",
    "df_actual_predicted_probs['Cumulative N Bad'] = df_actual_predicted_probs['Cumulative N Population'] - df_actual_predicted_probs['loan_data_targets_test'].cumsum()\n",
    "# We calculate cumulative number of 'bad', which is\n",
    "# the difference between the cumulative number of all observations and cumulative number of 'good' for each row."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs['Cumulative Perc Population'] = df_actual_predicted_probs['Cumulative N Population'] / (df_actual_predicted_probs.shape[0])\n",
    "\n",
    "df_actual_predicted_probs['Cumulative Perc Good'] = df_actual_predicted_probs['Cumulative N Good'] / df_actual_predicted_probs['loan_data_targets_test'].sum()\n",
    "\n",
    "df_actual_predicted_probs['Cumulative Perc Bad'] = df_actual_predicted_probs['Cumulative N Bad'] / (df_actual_predicted_probs.shape[0] - df_actual_predicted_probs['loan_data_targets_test'].sum())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Plot Gini\n",
    "plt.plot(df_actual_predicted_probs['Cumulative Perc Population'], df_actual_predicted_probs['Cumulative Perc Bad'])\n",
    "# We plot the cumulative percentage of all along the x-axis and the cumulative percentage 'good' along the y-axis,\n",
    "# thus plotting the Gini curve.\n",
    "plt.plot(df_actual_predicted_probs['Cumulative Perc Population'], df_actual_predicted_probs['Cumulative Perc Population'], linestyle = '--', color = 'k')\n",
    "# We plot a seconary diagonal line, with dashed line style and black color.\n",
    "plt.xlabel('Cumulative % Population')\n",
    "# We name the x-axis \"Cumulative % Population\".\n",
    "plt.ylabel('Cumulative % Bad')\n",
    "# We name the y-axis \"Cumulative % Bad\".\n",
    "plt.title('Gini')\n",
    "# We name the graph \"Gini\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Gini = AUROC * 2 - 1\n",
    "\n",
    "Gini"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot KS\n",
    "plt.plot(df_actual_predicted_probs['y_hat_test_proba'], df_actual_predicted_probs['Cumulative Perc Bad'], color = 'r')\n",
    "# We plot the predicted (estimated) probabilities along the x-axis and the cumulative percentage 'bad' along the y-axis,\n",
    "# colored in red.\n",
    "plt.plot(df_actual_predicted_probs['y_hat_test_proba'], df_actual_predicted_probs['Cumulative Perc Good'], color = 'b')\n",
    "# We plot the predicted (estimated) probabilities along the x-axis and the cumulative percentage 'good' along the y-axis,\n",
    "# colored in red.\n",
    "plt.xlabel('Estimated Probability for being Good')\n",
    "# We name the x-axis \"Estimated Probability for being Good\".\n",
    "plt.ylabel('Cumulative %')\n",
    "# We name the y-axis \"Cumulative %\".\n",
    "plt.title('Kolmogorov-Smirnov')\n",
    "# We name the graph \"Kolmogorov-Smirnov\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "KS = max(df_actual_predicted_probs['Cumulative Perc Bad'] - df_actual_predicted_probs['Cumulative Perc Good'])\n",
    "\n",
    "KS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.options.display.max_columns = None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ref_categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ref_categories = pd.DataFrame(ref_categories, columns = ['Feature name'])\n",
    "# We create a new dataframe with one column. Its values are the values from the 'reference_categories' list.\n",
    "# We name it 'Feature name'.\n",
    "df_ref_categories['Coefficients'] = 0\n",
    "# We create a second column, called 'Coefficients', which contains only 0 values.\n",
    "df_ref_categories['p_values'] = np.nan\n",
    "# We create a third column, called 'p_values', with contains only NaN values.\n",
    "df_ref_categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard = pd.concat([summary_table, df_ref_categories])\n",
    "\n",
    "df_scorecard = df_scorecard.reset_index()\n",
    "\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard['Original feature name'] = df_scorecard['Feature name'].str.split(':').str[0]\n",
    "\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "min_score = 300\n",
    "max_score = 850"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard.groupby('Original feature name')['Coefficients'].min()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "min_sum_coef = df_scorecard.groupby('Original feature name')['Coefficients'].min().sum()\n",
    "\n",
    "min_sum_coef"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "df_scorecard.groupby('Original feature name')['Coefficients'].max()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_sum_coef = df_scorecard.groupby('Original feature name')['Coefficients'].max().sum()\n",
    "\n",
    "max_sum_coef"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard['Score - Calculation'] = df_scorecard['Coefficients'] * (max_score - min_score) / (max_sum_coef - min_sum_coef)\n",
    "\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard['Score - Calculation'][0] = ((df_scorecard['Coefficients'][0] - min_sum_coef) / (max_sum_coef - min_sum_coef)) * (max_score - min_score) + min_score\n",
    "\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard['Score - Preliminary'] = df_scorecard['Score - Calculation'].round()\n",
    "\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "min_sum_score_prel = df_scorecard.groupby('Original feature name')['Score - Preliminary'].min().sum()\n",
    "\n",
    "min_sum_score_prel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_sum_score_prel = df_scorecard.groupby('Original feature name')['Score - Preliminary'].max().sum()\n",
    "\n",
    "max_sum_score_prel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard['Difference'] = df_scorecard['Score - Preliminary'] - df_scorecard['Score - Calculation']\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard['Score - Final'] = df_scorecard['Score - Preliminary']\n",
    "df_scorecard['Score - Final'][77] = 16\n",
    "df_scorecard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "min_sum_score_prel = df_scorecard.groupby('Original feature name')['Score - Final'].min().sum()\n",
    "\n",
    "min_sum_score_prel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "max_sum_score_prel = df_scorecard.groupby('Original feature name')['Score - Final'].max().sum()\n",
    "\n",
    "max_sum_score_prel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat_w_intercept = inputs_test_with_ref_cat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat_w_intercept.insert(0, 'Intercept', 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat_w_intercept.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat_w_intercept = inputs_test_with_ref_cat_w_intercept[df_scorecard['Feature name'].values]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat_w_intercept.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scorecard_scores = df_scorecard['Score - Final']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_test_with_ref_cat_w_intercept.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "scorecard_scores = scorecard_scores.values.reshape(102, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_scores = inputs_test_with_ref_cat_w_intercept.dot(scorecard_scores)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sum_coef_from_score = ((y_scores - min_score) / (max_score - min_score)) * (max_sum_coef - min_sum_coef) + min_sum_coef\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_proba_from_score = np.exp(sum_coef_from_score) / (np.exp(sum_coef_from_score) + 1)\n",
    "\n",
    "y_hat_proba_from_score.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat_test_proba[0: 5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_actual_predicted_probs['y_hat_test_proba'].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tr = 0.9\n",
    "df_actual_predicted_probs['y_hat_test'] = np.where(df_actual_predicted_probs['y_hat_test_proba'] > tr, 1, 0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted']) / df_actual_predicted_probs.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted']) / df_actual_predicted_probs.shape[0]).iloc[0, 0] + (pd.crosstab(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test'], rownames = ['Actual'], colnames = ['Predicted']) / df_actual_predicted_probs.shape[0]).iloc[1, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "roc_curve(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test_proba'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fpr, tpr, thresholds = roc_curve(df_actual_predicted_probs['loan_data_targets_test'], df_actual_predicted_probs['y_hat_test_proba'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(fpr, tpr)\n",
    "plt.plot(fpr, fpr, linestyle = '--', color = 'k')\n",
    "plt.xlabel('False positive rate')\n",
    "plt.ylabel('True positive rate')\n",
    "plt.title('ROC curve')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs = pd.concat([pd.DataFrame(thresholds), pd.DataFrame(fpr), pd.DataFrame(tpr)], axis = 1)\n",
    "# We concatenate 3 dataframes along the columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs.columns = ['thresholds', 'fpr', 'tpr']\n",
    "# We name the columns of the dataframe 'thresholds', 'fpr', and 'tpr'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs['thresholds'][0] = 1 - 1 / np.power(10, 16)\n",
    "# Let the first threshold (the value of the thresholds column with index 0) be equal to a number, very close to 1\n",
    "# but smaller than 1, say 1 - 1 / 10 ^ 16."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs['Score'] = ((np.log(df_cutoffs['thresholds'] / (1 - df_cutoffs['thresholds'])) - min_sum_coef) * ((max_score - min_score) / (max_sum_coef - min_sum_coef)) + min_score).round()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs['Score'][0] = max_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def n_approved(p):\n",
    "    return np.where(df_actual_predicted_probs['y_hat_test_proba'] >= p, 1, 0).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs['N Approved'] = df_cutoffs['thresholds'].apply(n_approved)\n",
    "\n",
    "df_cutoffs['N Rejected'] = df_actual_predicted_probs['y_hat_test_proba'].shape[0] - df_cutoffs['N Approved']\n",
    "\n",
    "df_cutoffs['Approval Rate'] = df_cutoffs['N Approved'] / df_actual_predicted_probs['y_hat_test_proba'].shape[0]\n",
    "\n",
    "df_cutoffs['Rejection Rate'] = 1 - df_cutoffs['Approval Rate']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs.iloc[5000: 6200, ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_cutoffs.iloc[1000: 2000, ]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs_train_with_ref_cat.to_csv('inputs_train_with_ref_cat.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_scorecard.to_csv('df_scorecard.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}