Article Sidebar

Download
PDF
Statistic
Read Counter : 277 Download : 164

Main Article Content

Abstract

The mechanism of switching and the identification of structural breakpoints in financial markets across different macro environments have long been core challenges in asset pricing and risk management. Traditional parametric models suffer from insufficient flexibility to capture nonlinear dynamic processes. This study proposes a regime-switching model driven by deep learning. By integrating a bidirectional long short-term memory network with the probabilistic inference framework of the Markov transformation process, a unified optimization framework for structural break identification, market regime classification, and volatility clustering modeling is constructed, in which the attention mechanism dynamically focuses on key historical information to enable mechanism identification. A multi-layer perceptron generates state-dependent GARCH parameters to adaptively capture the characteristics of heterogeneous fluctuations, and adaptive threshold monitoring based on KL divergence enables quantitative identification of structural breaks. Experiments show that the model achieves significant performance advantages over traditional methods in structural break detection, mechanism transition identification, and volatility prediction. The cross-market generalization ability and robustness analysis verify the model's applicability across different asset classes and time horizons. The posterior probability distribution of the model's state output can support asset allocation decisions, and the breakpoint identification mechanism provides quantitative early-warning indicators for regulators. It has important practical value in scenarios such as portfolio management, market timing strategies, and systemic risk prevention.

Keywords

Deep learning Regime switching model Structural break Volatility clustering Financial time series

Article Details

How to Cite
Wang, J. (2026). Deep learning-driven regime switching models for capturing structural breaks and volatility clustering in financial time series . Future Technology, 5(2), 268–280. Retrieved from https://fupubco.com/futech/article/view/815
Download Citation
Endnote/Zotero/Mendeley (RIS)
BibTeX
Bookmark and Share

References

  1. M. Chocholatá, "Volatility regimes of selected central European stock returns: a Markov switching GARCH approach," Journal of Business Economics and Management, vol. 23, no. 4, pp. 876–894-876–894, 2022. http://dx.doi.org/10.3846/jbem.2022.16648
  2. S. G. Hall, G. S. Tavlas, L. Trapani, and Y. Wang, "On the detection of structural breaks: the case of the Covid shock," Journal of Forecasting, vol. 44, no. 3, pp. 1042-1070, 2025. http://dx.doi.org/10.1002/for.3238
  3. M. Segnon, R. Gupta, and B. Wilfling, "Forecasting stock market volatility with regime-switching GARCH-MIDAS: The role of geopolitical risks," International Journal of Forecasting, vol. 40, no. 1, pp. 29-43, 2024. http://dx.doi.org/10.1016/j.ijforecast.2022.11.007
  4. B. Zhang, "A study of financial time series volatility forecasting method based on GARCH modeling," in Proceedings of the 2025 International Conference on Digital Economy and Intelligent Computing, 2025, pp. 54-59. http://dx.doi.org/10.1145/3746972.3746982
  5. O. V. De la Torre-Torres, D. Aguilasocho-Montoya, and M. d. l. C. del Río-Rama, "A two-regime Markov-switching GARCH active trading algorithm for coffee, cocoa, and sugar futures," Mathematics, vol. 8, no. 6, p. 1001, 2020. http://dx.doi.org/10.3390/math8061001
  6. I. Palupi, B. A. Wahyudi, and A. P. Putra, "Implementation of hidden Markov model (HMM) to predict financial market regime," in 2021 9th international conference on information and communication technology (ICOICT), 2021, pp. 639-644: IEEE. http://dx.doi.org/10.1109/ICoICT52021.2021.9527459
  7. M. J. Fülle, H. Herwartz, and S. Wang, "Markov-Switching Multivariate GARCH Model with Copula-Distributed Innovations," Available at SSRN 4984390, 2024. http://dx.doi.org/10.2139/ssrn.4984390
  8. L. Oelschläger, T. Adam, and R. Michels, "fHMM: Hidden Markov models for financial time series in R," Journal of Statistical Software, vol. 109, pp. 1-25, 2024. http://dx.doi.org/10.18637/jss.v109.i09
  9. F. Yin, Y. You, T. Wang, and M. Yu, "Pricing VIX Futures Under a Markov‐Switching GARCH Framework," Journal of Futures Markets, 2025. http://dx.doi.org/10.1002/fut.70041
  10. K. Sako, B. N. Mpinda, and P. C. Rodrigues, "Neural networks for financial time series forecasting," Entropy, vol. 24, no. 5, p. 657, 2022. http://dx.doi.org/10.3390/e24050657
  11. R. Qi and L. Dong, "Financial Time Series Forecasting Algorithm Based on Recurrent Neural Network," Advances in Artificial Intelligence, Big Data and Algorithms, pp. 339-345, 2023. http://dx.doi.org/10.3233/FAIA230828
  12. Z. Zeng, R. Kaur, S. Siddagangappa, S. Rahimi, T. Balch, and M. Veloso, "Financial time series forecasting using cnn and transformer," arXiv preprint arXiv:2304.04912, 2023. http://dx.doi.org/10.48550/arXiv.2304.04912
  13. A. Hadizadeh, M. J. Tarokh, and M. M. Ghazani, "A novel transformer-based dual attention architecture for the prediction of financial time series," Journal of King Saud University Computer and Information Sciences, vol. 37, no. 5, p. 72, 2025. http://dx.doi.org/10.1007/s44443-025-00045-y
  14. M. R. Kabir, D. Bhadra, M. Ridoy, and M. Milanova, "LSTM–transformer-based robust hybrid deep learning model for financial time series forecasting," Sci, vol. 7, no. 1, p. 7, 2025. http://dx.doi.org/10.3390/sci7010007
  15. M. Kolambe and S. Arora, "Time Series Forecasting Enhanced by Integrating GRU and N-BEATS," INTERNATIONAL JOURNAL OF INFORMATION, vol. 17, no. 1, pp. 140-158, 2025. http://dx.doi.org/10.5815/ijieeb.2025.01.07
  16. J. Ditzen, Y. Karavias, and J. Westerlund, "Testing and estimating structural breaks in time series and panel data in Stata," The Stata Journal, vol. 25, no. 3, pp. 526-560, 2025. http://dx.doi.org/10.48550/arXiv.2110.14550
  17. J. Ditzen, Y. Karavias, and J. Westerlund, "Multiple structural breaks in interactive effects panel data models," Journal of Applied Econometrics, vol. 40, no. 1, pp. 74-88, 2025. http://dx.doi.org/10.1002/jae.3097
  18. Z. Wang et al., "Revisiting vae for unsupervised time series anomaly detection: A frequency perspective," in Proceedings of the ACM web conference 2024, 2024, pp. 3096-3105. http://dx.doi.org/10.1145/3589334.3645710
  19. Z. Xu, J. Liechty, S. Benthall, N. Skar-Gislinge, and C. McComb, "GARCH-Informed Neural Networks for Volatility Prediction in Financial Markets," in Proceedings of the 5th ACM International Conference on AI in Finance, 2024, pp. 600-607. http://dx.doi.org/10.1145/3677052.3698600
  20. H. J. Hortúa and A. Mora-Valencia, "Forecasting VIX using Bayesian deep learning," International Journal of Data Science and Analytics, vol. 20, no. 3, pp. 2039-2060, 2025. http://dx.doi.org/10.1007/s41060-024-00562-5
  21. A. Botte and D. Bao, "A machine learning approach to regime modeling," Two Sigma, 2021.https://www.twosigma.com/articles/a-machine-learning-approach-to-regime-modeling/
  22. V. García, T. Blanco, and J. S. Sánchez, "A survey on uncertainty quantification in deep learning for financial time series prediction," Instituto de Ingeniería y Tecnología, 2024. http://dx.doi.org/10.1016/j.neucom.2024.127339
  23. X. Xu, H. Peng, and Y. Chen, "Deep Switching State Space Model for Nonlinear Time Series Forecasting with Regime Switching," arXiv preprint arXiv:2106.02329, 2021. http://dx.doi.org/10.1016/j.ijforecast.2025.05.001
  24. C. Mari and E. Mari, "Deep learning based regime-switching models of energy commodity prices," Energy Systems, vol. 14, no. 4, pp. 913-934, 2023. http://dx.doi.org/10.1007/s12667-022-00515-6
  25. I. Agakishiev, W. K. Härdle, D. Becker, and X. Zuo, "Regime switching forecasting for cryptocurrencies," Digital Finance, vol. 7, no. 1, pp. 107-131, 2025. http://dx.doi.org/10.1007/s42521-024-00123-2
  26. J. Wei, S. Yang, and Z. Cui, "Unified GARCH-Recurrent Neural Networks in Financial Volatility Forecasting," 2025.https://arxiv.org/html/2504.09380v2
  27. A. Chatterjee, H. Bhowmick, and J. Sen, "Stock volatility prediction using time series and deep learning approach," in 2022 IEEE 2nd Mysore Sub Section International Conference (MysuruCon), 2022, pp. 1-6: IEEE. http://dx.doi.org/10.1109/MysuruCon55714.2022.9972559
  28. H. T. Araya, J. Aduda, and T. Berhane, "A hybrid garch and deep learning method for volatility prediction," Journal of Applied Mathematics, vol. 2024, no. 1, p. 6305525, 2024. http://dx.doi.org/10.1155/2024/6305525
  29. G. Di-Giorgi, R. Salas, R. Avaria, C. Ubal, H. Rosas, and R. Torres, "Volatility forecasting using deep recurrent neural networks as GARCH models," Computational Statistics, vol. 40, no. 6, 2025. http://dx.doi.org/10.1007/s00180-023-01349-1
  30. S. An, X. Gao, F. An, and T. Wu, "Early warning of regime switching in a complex financial system from a spillover network dynamic perspective," iScience, vol. 28, no. 3, 2025. http://dx.doi.org/10.1016/j.isci.2025.111924
  31. A. Dezhkam, M. T. Manzuri, A. Aghapour, A. Karimi, A. Rabiee, and S. M. Shalmani, "A Bayesian-based classification framework for financial time series trend prediction," The Journal of supercomputing, vol. 79, no. 4, pp. 4622-4659, 2023. http://dx.doi.org/10.1007/s11227-022-04834-4
  32. O. B. Akgun and E. Gulay, "Dynamics in Realized Volatility Forecasting: Evaluating GARCH Models and Deep Learning Algorithms Across Parameter Variations," Computational Economics, vol. 65, no. 6, pp. 3971-4013, 2025. http://dx.doi.org/10.1007/s10614-024-10694-2
  33. Z. Wu, Q. Zhang, J. Zhou, H. Chen, and Y. Liu, "WAVAE: A Weakly Augmented Variational Autoencoder for Time Series Anomaly Detection," Information Fusion, p. 103462, 2025. http://dx.doi.org/10.1016/j.inffus.2025.103462