Modelling failure rates with machine-learning models: Evidence from a panel of UK firms

Sermpinis, G. , Tsoukas, S. and Zhang, Y. (2023) Modelling failure rates with machine-learning models: Evidence from a panel of UK firms. European Financial Management, 29(3), pp. 734-763. (doi: 10.1111/eufm.12369)

Abstract

In this study we investigate the ability of machine-learning techniques to predict firm failures, and we compare them against alternatives. Using data on business and financial risks on UK firms over 1994-2019, we document that machine-learning models are systematically more accurate than a discrete hazard benchmark. We conclude that the random forest model outperforms other models in failure prediction. In addition, we show that the improved predictive power of the random forest model relative to its counterparts, persists when we consider extreme economic events as well as firm and industry heterogeneity. Finally, we find that financial factors affect failure probabilities.

• Item Type: Articles
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Tsoukas, Professor Serafeim and Sermpinis, Professor Georgios
• Authors: Sermpinis, G., Tsoukas, S., and Zhang, Y.
• Subjects: H Social Sciences > HG Finance
• College/School: College of Social Sciences > Adam Smith Business School > Accounting and Finance
• Journal Name: European Financial Management
• Publisher: Wiley
• ISSN: 1354-7798
• ISSN (Online): 1468-036X
• Published Online: 05 May 2022
• Copyright Holders: Copyright © 2022 The Authors
• First Published: First published in European Financial Management 29(3):734-763
• Publisher Policy: Reproduced under a Creative Commons License