PSO–GA-based adaptive Kalman filter for estimation of market microstructure model
Discover Artificial Intelligence · 20.05.2026 · 6 (1)
DOI: 10.1007/s44163-026-01349-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Discover Artificial Intelligence, online erschienen am 20.05.2026. Verfasst von Honghui Xi und 3 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Xianda College of Economics & Humanities Shanghai International Studies University, Shanghai, National Science Foundation of China.
Abstract (Originalsprache)
This paper proposes a hybrid PSO–GA–KF framework that integrates particle swarm optimization (PSO) and a genetic algorithm (GA) within a likelihood-based Kalman filtering scheme for parameter estimation in Market Microstructure (MM) models. State estimation in MM models is sensitive to structural parameters, initial states, and noise covariance matrices. To address this issue, a hybrid PSO–GA strategy is employed to optimize the parameters of the Kalman filter. The proposed approach replaces the standard GA mutation operator with a PSO-inspired directed mutation mechanism, promoting guided exploration while maintaining population diversity. Simulation experiments and empirical analyses based on the Shenzhen Component Index (SZCI) and Hang Seng Index (HSI) demonstrate that the hybrid framework achieves improved estimation accuracy compared with standalone PSO–KF and GA–KF methods. The estimated excess demand closely tracks market price dynamics, while the corresponding liquidity estimates provide economically interpretable measures of trading conditions. An asset allocation application further illustrates the practical relevance of the proposed framework for dynamic trade timing and risk management. These results suggest that hybrid evolutionary optimization provides an effective approach for enhancing state-space modeling in financial market analysis.
Förderung
- Xianda College of Economics & Humanities Shanghai International Studies University, Shanghai
- National Science Foundation of China
Zitieren
APA
Xi, H., Xi, Y., Qin, Y., & Peng, H. (2026). PSO–GA-based adaptive Kalman filter for estimation of market microstructure model. Discover Artificial Intelligence, 6(1). https://doi.org/10.1007/s44163-026-01349-2
Vancouver
Xi H, Xi Y, Qin Y, Peng H. PSO–GA-based adaptive Kalman filter for estimation of market microstructure model. Discover Artificial Intelligence. 2026;6(1). doi:10.1007/s44163-026-01349-2
BibTeX
@article{Xi2026,
author = {Xi, Honghui and Xi, Yanhui and Qin, Yemei and Peng, Hui},
title = {PSO–GA-based adaptive Kalman filter for estimation of market microstructure model},
journal = {Discover Artificial Intelligence},
year = {2026},
volume = {6},
number = {1},
doi = {10.1007/s44163-026-01349-2}
}
EUVIMED