A Study on Quantitative Trading Strategies Based on Multimodal Data Fusion and Meta-Reinforcement Learning

Bolai Chong

doi:10.47297/wspdecWSP2515-797305.20250906

PDF
Export
Share
Collection
Album

A Study on Quantitative Trading Strategies Based on Multimodal Data Fusion and Meta-Reinforcement Learning

Volume 9, Issue 6

Views: 0 Downloads: 0

A Study on Quantitative Trading Strategies Based on Multimodal Data Fusion and Meta-Reinforcement Learning
A Study on Quantitative Trading Strategies Based on Multimodal Data Fusion and Meta-Reinforcement Learning
Department of Tourism and Hotel Management, Xiamen University, Xiamen, Fujian, 361005, China
Vol. 9, Issue 6, Pages: 25-28(2025)
Published：30 June 2025
DOI：10.47297/wspdecWSP2515-797305.20250906
Accepted：

Scan QR Code

Chong Bolai.A Study on Quantitative Trading Strategies Based on Multimodal Data Fusion and Meta-Reinforcement Learning[J].Development Economics of China,2025,09(06):25-28.
DOI：

Chong Bolai.A Study on Quantitative Trading Strategies Based on Multimodal Data Fusion and Meta-Reinforcement Learning[J].Development Economics of China,2025,09(06):25-28. DOI： 10.47297/wspdecWSP2515-797305.20250906.

摘要

Abstract

Quantitative trading often faces challenges such as strategy failure

insufficient data penetration

and limited path adaptability under extreme market conditions. This paper proposes a trading framework that integrates multimodal data—including news sentiment and stock price sequences—with meta-reinforcement learning to systematically enhance the adaptability and robustness of trading strategies. By constructing a hierarchical reinforcement learning system and a multimodal feature fusion network

the framework enables efficient perception and responsive scheduling under complex market dynamics. Experimental results demonstrate that the proposed method achieves superior risk-return performance in both U.S. and Chinese stock markets

with an average return improvement of over 15%. Furthermore

it maintains low volatility and maximum drawdown even under extreme policy shock scenarios

validating its robustness across markets and modalities.

关键词

Keywords

references

FOURMOU M . Algorithmic high-frequency trading: a systematic literature review [R/OL ] . 2024 .

ZHANG L , HUA L . Major issues in high-frequency financial data analysis: a survey of solutions [J ] . Mathematics , 2023 , 13 ( 3 ): 347 .

QIN M , SUN S , ZHANG W , et al . EarnHFT: efficient hierarchical reinforcement learning for high frequency trading [C/OL ] // Proceedings of the ... Conference . arXiv: 2309.12891 , 2023 .

LIU J . Multimodal data-driven factor models for stock market forecasting [J ] . Journal of Computer Technology and Software , 2025 , 4 ( 2 ): 17 - 23 .

DING X , ZHANG Y , LIU T , et al . Deep learning for event-driven stock prediction [C ] // Proceedings of the 24th International Joint Conference on Artificial Intelligence (IJCAI) . Buenos Aires, Argentina : AAAI Press , 2015 : 2327 - 2333 .

BOLLEN J , MAO H , ZENG X . Twitter mood predicts the stock market [J ] . Journal of Computational Science , 2011 , 2 ( 1 ): 1 - 8 .

XU Y , COHEN S B . Stock movement prediction from tweets and historical prices [C ] // Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL) . Melbourne, Australia : ACL , 2018 : 1970 - 1979 .

THEATE T , ERNST D . An application of deep reinforcement learning to algorithmic trading [J ] . Expert Systems with Applications , 2021 , 173 : 114632 .

JIANG Z , XU D , LIANG J . A deep reinforcement learning framework for the financial portfolio management problem [EB/OL ] . arXiv: 1706.10059 , 2017 .

TIAN Y , GAO M , GAO Q , et al . Trading in fast-changing markets with meta-reinforcement learning [J ] . Intelligent Automation & Soft Computing , 2024 , 39 ( 2 ): 175 - 188 .

CHEN Y Y , HUO J , DING T Y , et al . A survey of meta-reinforcement learning research [J ] . Journal of Software , 2024 , 35 ( 4 ): 1618 - 1650 .

KOSHIYAMA A , BLUMBERG S B , FIROOZYE N , et al . QuantNet: transferring learning across trading strategies [J ] . Quantitative Finance , 2022 , 22 ( 6 ): 1071 - 1090 .

LIU B , GAO X , YANG J , et al . FinRL: a deep reinforcement learning library for automated stock trading [EB/OL ] . arXiv: 2011.09607 , 2020 .

KUMAR S , VISHAL M , RAVI V . Explainable reinforcement learning on financial stock trading using SHAP [EB/OL ] . arXiv: 2208.08790 , 2022 .

Alert me when the article has been cited

Submit

No data

Related Author

No data

Related Institution

No data

⁰