Research on SZSE Component Index Volatility Prediction Based on CEEMDAN-BiLSTM
Journal: Modern Economics & Management Forum DOI: 10.32629/memf.v7i2.5179
Abstract
This study proposes a hybrid framework combining CEEMDAN and BiLSTM to predict high-frequency volatility of the SZSE Component Index. CEEMDAN first decomposes the volatility series into multi-scale components, which are then modeled individually by BiLSTM networks. Using 1-minute intraday data from 2014 to 2024, the proposed model significantly improves forecasting accuracy, achieving an R² of 0.9325 and reducing MAPE to 19.045%, substantially outperforming standard BiLSTM (R² = 0.5365, MAPE = 33.246%). Results demonstrate that signal decomposition effectively enhances the modeling of multi-scale volatility dynamics.
Keywords
CEEMDAN; Bidirectional LSTM; volatility prediction; SZSE Component Index
Full Text
PDF - Viewed/Downloaded: 3 TimesReferences
[1] Zhang, Y., Peng, Y. & Song, Y. Realized Volatility Forecasting for Stocks and Futures Indices with Rolling CEEMDAN and Machine Learning Models. Comput Econ 66, 1215–1268 (2025).
[2] Zeng, Q., Zhang, J., & Zhong, J. (2024). China's futures market volatility and sectoral stock market volatility prediction. Energy Economics, 132, 107429.
[3] Agnolucci, P. (2009). Volatility in crude oil futures: A comparison of the predictive ability of GARCH and implied volatility models. Energy Economics, 31(2), 316-321.
[4] Kim, H. Y., & Won, C. H. (2018). Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models. Expert Systems With Applications, 103, 25-37.
[5] Beniwal, M., Singh, A., & Kumar, N. (2023). Forecasting long-term stock prices of global indices: A forward-validating Genetic Algorithm optimization approach for Support Vector Regression. Applied Soft Computing, 145, 110566.
[6] Shen, J., & Shafiq, M. (2025). STL-ELM: A computationally efficient hybrid approach for predicting high volatility stock market. Scientific African, 28, e02590.
[7] Li, H., Mei, Y., Hao, X., & Chen, Z. (2024). Out-of-sample equity premium predictability: An EMD-denoising based model. Pacific-Basin Finance Journal, 88, 102536.
[2] Zeng, Q., Zhang, J., & Zhong, J. (2024). China's futures market volatility and sectoral stock market volatility prediction. Energy Economics, 132, 107429.
[3] Agnolucci, P. (2009). Volatility in crude oil futures: A comparison of the predictive ability of GARCH and implied volatility models. Energy Economics, 31(2), 316-321.
[4] Kim, H. Y., & Won, C. H. (2018). Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models. Expert Systems With Applications, 103, 25-37.
[5] Beniwal, M., Singh, A., & Kumar, N. (2023). Forecasting long-term stock prices of global indices: A forward-validating Genetic Algorithm optimization approach for Support Vector Regression. Applied Soft Computing, 145, 110566.
[6] Shen, J., & Shafiq, M. (2025). STL-ELM: A computationally efficient hybrid approach for predicting high volatility stock market. Scientific African, 28, e02590.
[7] Li, H., Mei, Y., Hao, X., & Chen, Z. (2024). Out-of-sample equity premium predictability: An EMD-denoising based model. Pacific-Basin Finance Journal, 88, 102536.
Copyright © 2026 Xin Yang
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License