Interpretable selective learning in credit risk

Dangxing Chen, Jiahui Ye and Weicheng Ye

Research in International Business and Finance, 2023, vol. 65, issue C

Abstract: Forecasting credit default risk has been an important research field for several decades. Traditionally, logistic regression has been widely recognized as a solution because of its accuracy and interpretability. Although complex machine learning models may improve accuracy over simple logistic regressions, their interpretability has prevented their use in credit risk assessment. We introduce a neural network with a selective option to increase interpretability by distinguishing whether linear models can explain the dataset. Our methods are tested on two datasets: 25,000 samples from the Taiwan payment system collected in October 2005 and 250,000 samples from the 2011 Kaggle competition. We find that, for most of samples, logistic regression will be sufficient, with reasonable accuracy; meanwhile, for some specific data portions, a shallow neural network model leads to much better accuracy without significantly sacrificing interpretability.

Keywords: Machine learning; Artificial intelligence; Interpretable learning; Credit risk; Risk assessment and management (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0275531923000661
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:riibaf:v:65:y:2023:i:c:s0275531923000661

DOI: 10.1016/j.ribaf.2023.101940

Access Statistics for this article

Research in International Business and Finance is currently edited by T. Lagoarde Segot

More articles in Research in International Business and Finance from Elsevier
Bibliographic data for series maintained by Catherine Liu ().