Credit Risk Modeling in the Age of Machine Learning

62 Pages Posted: 18 Nov 2021 Last revised: 25 Apr 2022

See all articles by Martin Thomas Hibbeln

Martin Thomas Hibbeln

University of Duisburg-Essen - Mercator School of Management

Raphael M. Kopp

University of Duisburg-Essen - Mercator School of Management

Noah Urban

University of Duisburg-Essen - Mercator School of Management

Date Written: April 24, 2022

Abstract

Based on the world’s largest loss database of corporate defaults, we perform a comparative analysis of machine learning (ML) methods in credit risk modeling across the globe. We find that ML methods, especially tree-based methods, substantially outperform both simple and sophisticated benchmarks. These results hold across different credit risk parameters, even though we use a uniform modeling framework for the ML methods. We find that the commonly applied out-of-sample validation—as opposed to out-of-time validation—results in inflated performance measures, consistent with an “information leakage channel” that induces information spillovers, particularly for macroeconomic features; this problem is prevalent in many economic contexts. Our results provide guidance for financial institutions, regulatory authorities, and academics.

Keywords: risk management, credit risk modeling, machine learning, forecasting, macroeconomic variables

JEL Classification: C18, C52, C53, C55, G17, G21

Suggested Citation

Hibbeln, Martin Thomas and Kopp, Raphael M. and Urban, Noah, Credit Risk Modeling in the Age of Machine Learning (April 24, 2022). Available at SSRN: https://ssrn.com/abstract=3913710 or http://dx.doi.org/10.2139/ssrn.3913710

Martin Thomas Hibbeln (Contact Author)

University of Duisburg-Essen - Mercator School of Management ( email )

Lotharstraße 65
Duisburg, Nordrhein-Westfalen 47057
Germany
+49 203 379-2830 (Phone)

Raphael M. Kopp

University of Duisburg-Essen - Mercator School of Management ( email )

Lotharstraße 65
Duisburg, Nordrhein-Westfalen 47057
Germany

Noah Urban

University of Duisburg-Essen - Mercator School of Management ( email )

Lotharstrasse 65
Duisburg, Nordrhein-Westfalen 47057
Germany

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
328
Abstract Views
896
Rank
149,945
PlumX Metrics