← Search

Depicting Risk Profile over Time: A Novel Multiperiod Loan Default Prediction Approach

Zhao Wang¹; Cuiqing Jiang²; Huimin Zhao³

¹ School of Management, Hefei University of Technology, The Philosophy and Social Sciences Key Laboratory of Digital Economy and Smart Enterprise Management, Anhui Province, Hefei, China · ² School of Management, Hefei University of Technology, The Philosophy and Social Sciences Laboratory of Data Science and Intelligent Society Governance, Ministry of Education, Hefei, China · ³ Sheldon B. Lubar College of Business, University of Wisconsin–Milwaukee, Milwaukee, WI, U.S.A.

With the rapid development of fintech, the need for dynamic credit risk evaluation is becoming increasingly important. While previous studies on credit scoring have mostly focused on single-period loan default prediction, we call for a new avenue—multiperiod default prediction (MPDP)—to depict risk profiles over time. To address the challenges raised by MPDP, such as monotonic default probability prediction and complex relationship accommodation, we propose a novel approach, hybrid and collective scoring (HACS). We design a hybrid modeling strategy to predict whether and when a borrower will default separately through a default discrimination model and a default time estimation model, respectively, and synthesize them through a probabilistic framework. To accommodate various possible patterns of default time and measure the distribution of default probability over successive time intervals, we propose a joint default modeling method to train the default time estimation model. Empirical evaluations at the model (time-to-default prediction performance and discrimination performance) and mechanism (identifiability and discriminability) levels, as well as impact analyses at the application (granting performance and profitability performance) level, show that HACS outperforms the benchmarked survival analysis and multilabel learning methods on all fronts. It can more accurately predict time-to-default and provide financial institutions and investors better decision-support in granting loans and selecting loan portfolios.

DOI

10.25300/misq/2022/17491

Volume

47 (4)

Pages

1455-1486

Language

en

Export

BibTeX

Sources

crossref