Abstract
Although deep reinforcement learning for portfolio optimization has attracted the attention of more and more researchers, existing research focuses on the improvement of a single algorithm. According to No Free Lunch Theorem (NFL), single algorithms are always limited, especially in complex financial environment. In this paper, an ensemble strategy that combines the advantages of three deep reinforcement learning algorithms is proposed to select appropriate agents at different stages. The ensemble strategy is composed of Deep Deterministic Policy Gradient (DDPG), Soft Actor-Critic (SAC) and Twin Delayed Deep Deterministic Policy Gradient (TD3). Compared with other model and algorithms, our model on experimental datasets shows better performance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Zhang, Y., Zhao, P., Wu, Q., et al.: Cost-sensitive portfolio selection via deep reinforcement learning. IEEE Trans. Knowl. Data Eng. 34(1), 236–248 (2022)
Tianxin, L., Xiaoping, Y., Liang, W., et al.: Research and development of financial trading system based on reinforcement learning. J. Softw. 30(03), 845–864 (2019)
Bing, Y., Ting, L., Jian, X., et al.: Deep reinforcement learning based on transformer and U-Net framework for stock trading. Knowl.-Based Syst. 262, 1–10 (2023)
Tianxiang, C., Shusheng, D., Huan, J., et al.: Portfolio constructions in cryptocurrency market: A CVaR-based deep reinforcement learning approach. Econ. Model. 119, 1–9 (2023)
Giulio, M., Raffaele, M.: Shrinkage estimation with reinforcement learning of large variance matrices for portfolio selection. Intell. Syst. Appl. 17, 1–11 (2023)
Jie, X., Yukun, Z., Chunxiao, X.: Research on financial trading algorithms based on deep reinforcement learning. Comput. Eng. Appl. 58(07), 276–285 (2022)
Hyungjun, P., Kyu, S.M., Gu, C.D.: An intelligent financial portfolio trading strategy using deep Q-learning. Expert Syst. Appl. 158, 1–16 (2020)
Mohamed, A.A., Xu, Z., Lee, C.-G.: What is the value of the cross-sectional approach to deep reinforcement learning? Quantitative Finan. 22(6), 1091–1111 (2022)
Tuomas, H., Aurick, Z., Pieter, A., et al.: Soft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor. In: Proceedings of the 35th International Conference on Machine Learning, pp. 1861–1870. PMLR (2018)
Lillicrap Timothy, P., Hunt Jonathan, J., Alexander, P., et al.: Continuous control with deep reinforcement learning. In: International Conference on Learning Representations (2016)
Scott, F., van Herke, H., David, M.: Addressing function approximation error in actor-critic methods. In: Proceedings of the 35th International Conference on Machine Learning, pp. 1587–1596. PMLR (2018)
Yang, H., Liu, X.-Y., Zhong, S., et al.: Deep reinforcement learning for automated stock trading: an ensemble strategy. In: Proceedings of the First ACM International Conference on AI in Finance, pp. 1–8. Association for Computing Machinery, New York (2020)
Acknowledgements
This work was supported by the Natural Science Foundation of Guangdong Province (2021A1515012298), the Project of Philosophy and Social Science Planning of Guangdong (GD21YGL16).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Su, X., Zhou, Y., He, S., Li, X. (2023). Ensemble Strategy Based on Deep Reinforcement Learning for Portfolio Optimization. In: Jin, Z., Jiang, Y., Buchmann, R.A., Bi, Y., Ghiran, AM., Ma, W. (eds) Knowledge Science, Engineering and Management. KSEM 2023. Lecture Notes in Computer Science(), vol 14120. Springer, Cham. https://doi.org/10.1007/978-3-031-40292-0_20
Download citation
DOI: https://doi.org/10.1007/978-3-031-40292-0_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-40291-3
Online ISBN: 978-3-031-40292-0
eBook Packages: Computer ScienceComputer Science (R0)