As an international consumer finance provider, Home Credit offers affordable point-of-sale loan, cash loan and revolving loan product to its clients in 9 countries. With over 100M customers served under 434K distribution points, Home Credit strives to provide great and reliable financial services to its main customers, underserved borrowers with limited credit history.
Currently, Home Credit relies on applicants’ external credit history with statistical and machine learning methods to drive lending decisions. While existing model proves to be effective, Home Credit wants to further discover hidden traits in their data to unlock its full potential. With the combination of each consumer’s Credit Bureau and previous Home Credit application data, our team conducted Logistic Regression, Random Forest, XGBoost and LightGBM to help the company develop an enhanced prediction model that would lead to more concise and intelligence conclusions. With the implementation of our prediction model, Home Credit can further expand financial inclusion for the unbanked population and create a win-win situation for both parties.