Hidden Markov Models for Analyzing U.S. Bank Failures

  • Authors

    • Sunil K Sapra California State University, Los Angeles
    https://doi.org/10.14419/zbqm4d90

    Received date: September 28, 2025

    Accepted date: November 3, 2025

    Published date: November 7, 2025

  • Hidden Markov Models; State Space Models; Regime Switching Models; Poisson HMM; EM Algorithm; Viterbi ‎Algorithm

  • Abstract

    This paper examines the use of Hidden Markov Models (HMMs) to investigate patterns in US bank failures. HMMs ‎are statistical tools designed for handling sequential data and uncovering hidden states, making them particularly ‎useful for studying systemic financial events. The paper provides an application of HMMs to U.S. bank failures ‎data and presents results of an empirical study that analyzes historical data on banking crises in the United States. ‎Four HMMs are employed for the analysis of U.S. bank failures data, and their relative performance is studied with ‎respect to model fitting and forecasting. Empirical results are presented for each of the HMMs employed. Global ‎decoding is employed to predict the most likely state sequence of large bank failure events via the Viterbi ‎algorithm. The findings demonstrate the ability of HMMs to uncover unobservable economic conditions and ‎enhance predictive capabilities for financial stability evaluations.

    Author Biography

    • Sunil K Sapra, California State University, Los Angeles

      Profesor, Department of Economics and Statistics, California State University, Los Angeles, USA

      Sunil Sapra, earned both his M.Phil. and Ph.D. in Econometrics at Columbia University, New York. He is currently a Professor of Economics and Statistics at CaliforniaStateUniversity, Los Angeles. Prior to joining Cal. State, LA, he taught at State University of New York, Buffalo and held the prestigious ASA/NSF/Census Research Fellowship (1989-90) at the Bureau of the Census, Washington, D. C. He is an expert in Business Statistics at the Westlaw Roundtable Group. He has published more than 70 articles in some of the most prestigious statistics and econometrics journals. His research on semi-parametric econometrics, missing data problems, nonlinear statistical and econometric models, robust statistical procedures, limited dependent variables and duration data analysis has been published in Econometric Theory, The American Statistician, International Journal of Advanced Statistics and Probability, Statistica Neerlandica, Statistical Papers, Sankhya, Economics e-journal, Bulletin of Economic Research, Economics Letters, Applied Economics Letters, and Empirical Economics Letters. He serves on the editorial boards of several statistics and economics journals. His research has been cited in statistics and econometrics textbooks and journals as well as a widely-used volume on statistical distributions. He received the Outstanding Professor Award from California State University, Los Angeles in 2003 for excellence in teaching and research. He has been listed in Who’s Who among American Teachers and Educators (11th Edition, 2007).

  • References

    1. Dempster, A.P., Laird, N.M. and Rubin, D., Maximum Likelihood from Incomplete Data via the EM Algorithm. Journal of the Royal Statistical So-ciety: Series B, 39, (1977),1-38. https://doi.org/10.1111/j.2517-6161.1977.tb01600.x.
    2. Hamilton, J. D., Time Series Analysis. Princeton University Press, (1994). https://doi.org/10.1515/9780691218632.
    3. Hamilton, James, Dates of U.S. recessions as inferred by GDP-based recession indicator [JHDUSRGDPBR], Retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/JHDUSRGDPBR, September 21, 2025.
    4. Michelot, Théo, HmmTMB: Hidden Markov Models with Flexible Covariate Effects in R.” arXiv Preprint arXiv:2211.14139, (2022).
    5. Michelot, Theo, Analysing time series data with hidden Markov models in hmmTMB, (2024).
    6. Rabiner, L. R., A tutorial on Hidden Markov Models and selected applications in speech recognition. Proceedings of the IEEE, 77(2), (1989), pp. 257–286. https://doi.org/10.1109/5.18626.
    7. Reiff, Nathan, Historical U.S. Unemployment Rate by Year, Retrieved from https://www.investopedia.com/historical-us-unemployment-rate-by-year-7495494, September 18, 2025.
    8. Sanchez-Espigares, Josep A., and Alberto Lopez-Moreno, R-package MSwM, (2021).
    9. Zucchini, Walter, Iain L MacDonald, and Roland Langrock, Hidden Markov Models for Time Series: An Introduction Using R, Second Edition. CRC, Boca Raton (2017). https://doi.org/10.1201/b20790.
    10. Bank Failures in Brief-Summary, Retrieved from FDIC: https://www.fdic.gov/resources/resolutions/bank-failures/in-brief/index, September 18, 2025.
    11. Board of Governors of the Federal Reserve System (US), Federal Funds Effective Rate [FEDFUNDS], Retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/FEDFUNDS, September 18, 2025.
    12. U.S. Bureau of Labor Statistics, Databases, Tables & Calculators by Subject. “Labor Force Statistics from the Current Population Survey: Unem-ployment Rate,” select years 2001 to 2024. Retrieved from Bureau of Labor Statistics Data, September 18, 2025.

  • Downloads