Abstract
Credit scoring is the risk assessment of customers. A reliable credit scoring model can provide decision support for financial institutions. In this paper, the logistic regression model with elastic net penalty (LR-EN) is proposed to assess personal credit score. Results on German bank personal credit data show that the proposed method can greatly improve classification precision of “bad” customer compared with other three methods. In addition, the attributes selected by LR-EN are well interpreted.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have