fitria-dwi / Credit-Risk-Prediction

This project was made as a final project at Rakamin Academy in collaboration with ID/X Partner.

Credit Risk Prediction

Created by Fitria Dwi Wulandari

Objectives:

1. Identified patterns that indicate if a person is unlikely to repay the loan or labeled as a bad risk.
2. Implemented machine learning algorithms to build a predictive model to predict loan risk from applicants.

Results:

1. The most important feature in determining whether the applicant has the possibility of not repaying the loan is the last payment month, the last payment amount, the principal amount received, the recovery value, and the last payment year.
2. The best model to predict the risk status of loan applications is Random Forest.

Documentation :

Click on this if you would like to go further to the notebook.

References:

• https://www.investopedia.com/terms/g/grace_period.asp
• https://www.investopedia.com/terms/d/default2.asp
• https://www.valuepenguin.com/loans/what-does-it-mean-to-default-on-a-loan
• https://www.bankrate.com/personal-finance/debt/charged-off-as-bad-debt/#:~:text=If%20you've%20been%20delinquent,a%20loss%20for%20the%20company.
• https://www.investopedia.com/terms/c/chargeoff.asp