Tail Risk Early Warning System for Capital Markets Based on Machine Learning Algorithms

Zongxin Zhang and Ying Chen ()
Additional contact information
Zongxin Zhang: Fudan University
Ying Chen: Fudan University

Computational Economics, 2022, vol. 60, issue 3, No 5, 923 pages

Abstract: Abstract Scientific and effective tail risk measurement and early warning are key points and difficulties in the identification and control of major risks in capital markets. In this paper, we use the autoregressive conditional Fréchet model (AcF) to construct a tail risk measurement index for the capital market in China. The tail risk status identified by the scientific index method is used as a monitoring anchor to construct and optimize a tail risk early warning model based on machine learning algorithms. The study yields three findings. (1) The AcF model can overcome the shortcomings of traditional models in tail risk measurement and significantly improve the tail risk measurement efficiency of the capital market. (2) Tail risk synergies between equity and bond markets are significantly stronger than yield synergies, and the tail risk measure index has the role of a leading indicator of significant risk in capital markets. (3) Based on the joint test of risk status and crisis identification efficiency, the Logit model of crisis identification fails whereas the tail risk warning model optimized by machine learning algorithms can accurately identify crises and significant risks. The optimal early warning model pairings for the stock market and bond market are the oversampling-random forest algorithm and the double sampling-random forest algorithm, respectively, with out-of-sample crisis warning accuracies of 81.94% and 90.20%, respectively.

Keywords: Tail Risk Measurement; Risk Warning; AcF Model; Machine Learning Algorithms (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://link.springer.com/10.1007/s10614-021-10171-0 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:60:y:2022:i:3:d:10.1007_s10614-021-10171-0

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

DOI: 10.1007/s10614-021-10171-0

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 ().