Using Hidden Markov Model to Detect Macro-economic Risk Level

Zhu, Yajing; Cheng, Jie

Abstract

In this paper, inspired by Moody’s BET model, a stochastic hidden Markov model is constructed to detect the macro-economic risk states hidden in the corporate default data. The observed default statistics are from four geographic regions, namely Asia-Pacific, Europe, the U.S. and the globe as a whole. The EM algorithm is applied to estimate parameters in each model, where the associated standard errors are computed using the Monte Carlo method. The validity of the binomial distribution assumption is checked by conducting the Chi-square goodness-of-fit test. When compared with the historical recession and expansion periods, most of the estimated risk-switching processes are in accord with the actual fluctuations in the macro-economy.

Citation

Zhu, Y., & Cheng, J. (2013). Using Hidden Markov Model to Detect Macro-economic Risk Level. Review of Integrative Business and Economics Research (RIBER), 2(1), 238-249

• Journal Article Type: Article
• Publication Date: 2013
• Deposit Date: Dec 14, 2023
• Journal: Review of Integrative Business and Economics Research
• Print ISSN: 2414-6722
• Publisher: Society of Interdisciplinary Business Research
• Peer Reviewed: Peer Reviewed
• Volume: 2
• Issue: 1
• Pages: 238-249
• Keywords: hidden Markov model; credit default analysis; EM algorithm; Monte Carlo method.
• Publisher URL: https://buscompress.com/uploads/3/4/9/8/34980536/riber_k13-075__238-249_.pdf
• Related Public URLs: https://buscompress.com/riber-2-1.html