Understanding and Predicting Systemic Corporate Distress: A Machine-Learning Approach
IMF Working Papers, July 29, 2022
Source details
- Canonical URL
- Understanding and Predicting Systemic Corporate Distress: A Machine-Learning Approach
Other formats
Bibliographic details
- Authors: Burcu Hacibedel, Ritong Qu
- Published: July 29, 2022
- Series: IMF Working Papers
- DOI: https://doi.org/10.5089/9798400216299.001
Scope and Data
- Study covers firm-level probabilities of default (PD) across 55 economies.
- Time span: the last three decades.
- Systemic corporate distress is identified by elevated PDs across a large portion of the firms in an economy.
- A machine-learning based early warning system is constructed to predict the onset of distress in one year’s time.
Key Findings
- Predictors of systemic corporate distress:
- Credit expansion.
- Monetary policy tightening.
- Overvalued stock prices.
- Debt-linked balance-sheet weaknesses.
- Associations of systemic corporate distress events:
- Associated with contractions in GDP and credit growth in advanced and emerging markets.
- Contractions associated with distress are milder than those associated with financial crises.
Methodology and Components (as described)
- Use of firm-level PD indices to identify distress events.
- Machine-learning model used for one-year-ahead early warning.
- Appendices referenced in keywords: appendix B constructing predictor, appendix C machine learning model.
Publication and Metadata
- Authors: Burcu Hacibedel, Ritong Qu
- Date: July 29, 2022
- Pages: 48
- Series: Working Paper No. 2022/153
- Volume: 2022
- Issue: 153
- DOI: https://doi.org/10.5089/9798400216299.001
- Stock No: WPIEA2022153
- ISBN: 9798400216299
- ISSN: 1018-5941
Source: IMF Working Paper No. 2022/153 by Burcu Hacibedel and Ritong Qu, July 29, 2022.
Content in this bundle
- Working Paper