Abstract
We analyze consumer defaults in a sample of 64 000 customers taking personal loans from a Korean bank. Applying a generalized additive modeling (GAM) framework, we show a nonlinear impact of loan and borrower characteristics. In particular, the likelihood of default is high for both low-income borrowers and high-income borrow-ers. Our results are robust to a range of different tests, and they highlight the useful-ness of the GAM framework, especially the graphical presentation of nonlinearities.
Original language | English |
---|---|
Pages (from-to) | 77-103 |
Number of pages | 27 |
Journal | Journal of Credit Risk |
Volume | 18 |
Issue number | 3 |
Early online date | 26 Aug 2022 |
DOIs | |
Publication status | Published - Sep 2022 |
Keywords
- Generalised additive models; B-spline; credit scoring; loan defaults; signal detection theory; mis-classification costs
- misclassification costs
- generalized additive models
- signal detection theory
- credit scoring
- loan defaults
- basis splines (B-splines)