Can Machine Learning Models Capture Correlations in Corporate Distresses?

32 Pages Posted: 18 Nov 2018 Last revised: 13 Jul 2019

See all articles by Benjamin Christoffersen

Benjamin Christoffersen

Copenhagen Business School - Department of Finance

Rastin Matin

National Bank of Denmark

Pia Mølgaard

National Bank of Denmark

Date Written: July 12, 2019

Abstract

A number of papers document that recent machine learning models outperform traditional corporate distress models in terms of accurately ranking firms by their riskiness. However, it remains unanswered whether advanced machine learning models can capture correlations in distresses sufficiently well to be used for joint modelling, which traditional distress models often struggle with. We implement a regularly top-performing machine learning model and find that prediction accuracy of individual distress probabilities improves while there is almost no difference in the predicted aggregate distress rate relative to traditional distress models. Thus, our findings suggest that complex machine learning models do not eliminate the excess clustering in distresses. Instead, we propose a frailty model, which allows for correlations in distresses, augmented with regression splines. This model demonstrates competitive performance in terms of ranking firms by their riskiness, while providing accurate aggregate risk measures.

Keywords: corporate default prediction, discrete hazard models, frailty models, gradient boosting

JEL Classification: C55, G17, G33

Suggested Citation

Christoffersen, Benjamin and Matin, Rastin and Mølgaard, Pia, Can Machine Learning Models Capture Correlations in Corporate Distresses? (July 12, 2019). Available at SSRN: https://ssrn.com/abstract=3273985 or http://dx.doi.org/10.2139/ssrn.3273985

Benjamin Christoffersen (Contact Author)

Copenhagen Business School - Department of Finance ( email )

A4.17 Solbjerg Plads 3
Copenhagen, Frederiksberg 2000
Denmark

Rastin Matin

National Bank of Denmark ( email )

Havnegade 5
DK-1093 Copenhagen K
Denmark

Pia Mølgaard

National Bank of Denmark ( email )

Havnegade 5
DK-1093 Copenhagen K
Denmark

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

Paper statistics

Downloads
99
Abstract Views
879
rank
309,119
PlumX Metrics