TY - GEN
T1 - Dynamic Predicting for Nonstationary Financial Signal Based on Variational Mode Decomposition and Variational Autoencoder
AU - Zhang, Bowei
AU - Chen, Yunzhu
AU - Zhang, Wenyu
AU - Li, Yuqing
AU - Ye, Neng
AU - Li, Xiangming
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
PY - 2025
Y1 - 2025
N2 - In recent years, data-driven machine learning techniques have made significant contributions to asset pricing, which uses factor models to estimate croSS-sectional expected returns. However, learning effective models from non-stationary and noisy financial data still remains a challenge. In this paper, we use variational mode decomposition (VMD) to construct observable characteristics that measure the unobservable dynamic loadings from stock price-volume data. Furthermore, we adopt a conditional variational autoencoder (VAE) architecture to extract low-dimensional factors and time-varying loadings by introducing the characteristics. Compared with the standard FactorVAE, our two-stage framework can improve the RankICIR metric by 27.8%,which represents higher predictive accuracy. Empirical tests on Chinese stock market also confirm the efficiency of our method.
AB - In recent years, data-driven machine learning techniques have made significant contributions to asset pricing, which uses factor models to estimate croSS-sectional expected returns. However, learning effective models from non-stationary and noisy financial data still remains a challenge. In this paper, we use variational mode decomposition (VMD) to construct observable characteristics that measure the unobservable dynamic loadings from stock price-volume data. Furthermore, we adopt a conditional variational autoencoder (VAE) architecture to extract low-dimensional factors and time-varying loadings by introducing the characteristics. Compared with the standard FactorVAE, our two-stage framework can improve the RankICIR metric by 27.8%,which represents higher predictive accuracy. Empirical tests on Chinese stock market also confirm the efficiency of our method.
KW - Dynamic factor model
KW - Financial signal processing
KW - Stock market prediction
KW - VAE
KW - VMD
UR - https://www.scopus.com/pages/publications/105008938095
U2 - 10.1007/978-981-96-6468-9_31
DO - 10.1007/978-981-96-6468-9_31
M3 - Conference contribution
AN - SCOPUS:105008938095
SN - 9789819664672
T3 - Communications in Computer and Information Science
SP - 353
EP - 363
BT - Information Processing and Network Provisioning - 3rd International Conference, ICIPNP 2024, Proceedings
A2 - Kadoch, Michel
A2 - Cheriet, Mohamed
A2 - Qiu, Xuesong
PB - Springer Science and Business Media Deutschland GmbH
T2 - 3rd International Conference on Information Processing and Network Provisioning, ICIPNP 2024
Y2 - 14 June 2024 through 16 June 2024
ER -