logistic_regression.py

import pandas as pd 
import statsmodels.api as sm 
import matplotlib
matplotlib.use('TkAgg')
import matplotlib.pyplot as plt 
import numpy as np 
import sys


def logit_function(x):
	p = 1/(1 + np.exp(x))
	return p

def logistic_function(fico_sore, loan_amount, coeff):
	
	FICO_coeff = coeff['FICO.Score']
	loan_amount_coeff = coeff['Amount.Requested']
	intercept = coeff['constant']

	x = intercept + FICO_coeff*fico_score + loan_amount_coeff*loan_amount
	p = logit_function(x)

	return p


def data_prep():
	data = pd.read_csv('cleaned_loansData.csv')

	data['IR_TF'] = data['Interest.Rate'].map(lambda x: 0 if x<12 else 1)
	data['constant'] = float(1.0)

	ind_vars = ['FICO.Score', 'Amount.Requested', 'constant']

	return data, ind_vars


if __name__ == '__main__':
	if len(sys.argv) != 3:
		print("Please input a FICO Score and Loan Amount")
		sys.exit()
	else:
		try:
			fico_score = float(sys.argv[1])
			loan_amount = float(sys.argv[2])

			data, ind_vars = data_prep()

			logit = sm.Logit(data['IR_TF'], data[ind_vars])
			result = logit.fit()
			coeff = result.params

			p = logistic_function(fico_score, loan_amount, coeff)
			print('The probability that you will get a loan from Lending Club with a FICO score of ' + str(fico_score) + ' and a loan amount of $' +
				 str(loan_amount) + ' is: ' + str(p*100) + '%.')
			plt.figure(figsize=(6,6))
			plt.plot(p)

			plt.show()
		except ValueError:
			print("Please enter a number for both the FICO Score and Loan Amount")