Explaining Mortgage Defaults Using SHAP and LASSO

Belma Ozturkkal () and Ranik Raaen Wahlstrøm ()
Additional contact information
Belma Ozturkkal: Kadir Has University
Ranik Raaen Wahlstrøm: Norwegian University of Science and Technology

Computational Economics, 2025, vol. 66, issue 4, No 20, 3325 pages

Abstract: Abstract We utilize machine learning methods to model the credit risk of mortgages in a significant emerging market. For this purpose, we investigate a multitude of variables that explain the characteristics of the loans, the demographics of the borrowers, and macroeconomic factors. We employ SHapley Additive exPlanations (SHAP) values in conjunction with five different tree-based machine learning methods, as well as the least absolute shrinkage and selection operator (LASSO) in conjunction with logistic regressions. Our findings, which are robust across two sampling schemes, reveal that while demographic variables are significant and important, loan-specific and macroeconomic variables are the most crucial in explaining mortgage defaults. As existing literature on mortgage default has primarily focused on advanced markets, we aim to bridge this gap by concentrating on emerging market data. We also share our code, which we hope will encourage others to utilize the methods we have applied.

Keywords: Mortgage loan; Default risk; Tree-based methods; LASSO; SHAP; Emerging market (search for similar items in EconPapers)
JEL-codes: C40 C52 C65 G21 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10614-024-10763-6 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

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:kap:compec:v:66:y:2025:i:4:d:10.1007_s10614-024-10763-6

Ordering information: This journal article can be ordered from
http://www.springer. ... ry/journal/10614/PS2

DOI: 10.1007/s10614-024-10763-6

Access Statistics for this article

Computational Economics is currently edited by Hans Amman

More articles in Computational Economics from Springer, Society for Computational Economics Contact information at EDIRC.
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().