skip to main content
10.1145/3650215.3650362acmotherconferencesArticle/Chapter ViewAbstractPublication PagesicmlcaConference Proceedingsconference-collections
research-article

A novel portfolio strategy approach using deep reinforcement learning

Published: 16 April 2024 Publication History

Abstract

The problem of portfolio strategy is an enduring topic in the financial field. The combination of deep learning and reinforcement learning for portfolio problems and the purpose of achieving intelligent transactions is an important research direction in the information technology era. Based on deep reinforcement learning, this paper uses deep learning BiLSTM to predict the rise and fall of stock prices, so that the agents of reinforcement learning can observe and better judge the current situation, so as to determine their own trading actions, so as to generate the optimal portfolio strategy. In this paper, ten stocks of US stocks are selected for experiments. Under the real market simulation, it is shown that compared with the benchmark method, the cumulative return of the model based on the deep reinforcement learning algorithm reaches 87.4 %, the return rate is higher, the risk is the smallest, and it has certain practical value.

References

[1]
Heaton J B, Polson N G, Witte J H. 2017. Deep learning for finance: deep portfolios. Applied Stochastic Models in Business and Industry vol 33(Wiley Press) pp 3–12.
[2]
Lei K, Zhang B, Li Y. 2020. Time-driven feature-aware jointly deep reinforcement learning for financial signal representation and algorithmic trading. Expert Systems with Applications vol 140 (Elsevier press) p 112872.
[3]
Mnih V, Badia A P, Mirza M. 2016. Asynchronous methods for deep reinforcement learning. Int. Conf. on Machine Learning (New York, U.S.) pp 1928–1937.
[4]
Moody J E, Saffell M, Liao Y. 1998. Reinforcement Learning for Trading Systems and Portfolios. Int. Conf. on Knowledge Discovery and Data Mining (New York, U.S.) pp 279-283.
[5]
Deng Y, Bao F, Kong Y. 2016. Deep direct reinforcement learning for financial signal representation and trading. IEEE Transactions on Neural Networks and Learning Systems vol 28 (IEEE press) pp 653–664.
[6]
Jiang Z, Liang J. 2017. Cryptocurrency portfolio management with deep reinforcement learning. Int. Conf. Intelligent Systems (London, U.K./IEEE press) pp 905–913.
[7]
Liang Q, Zhu M, Zheng X. 2021. An Adaptive News-Driven Method for CVaR-Sensitive Online Portfolio Selection in Non-Stationary Financial Markets. Int. Joint Conf. on Artificial Intelligence. (Montreal, Canada) pp 2708–2715.
[8]
Gao Z, Gao Y, Hu Y. 2020. Application of deep q-network in portfolio management. Int. Conf. on Big Data Analytics (Atlanta, U.S.A/IEEE press) pp 268–275.
[9]
Weng L, Sun X, Xia M. 2020. Portfolio Trading System of Digital Currencies: A Deep Reinforcement Learning with Multidimensional Attention Gating Mechanism. Neurocomputing, vol 402 (Elsevier press) pp 171–182.
[10]
Ye Y, Pei H, Wang B. 2020. Reinforcement-learning based portfolio management with augmented asset movement prediction states Proceedings of the AAAI Conf. on Artificial Intelligence vol 4 (California, U.S.A./AAAI press) p 34.
[11]
Li L. 2021. An Automated Portfolio Trading System with Feature Preprocessing and Recurrent Reinforcement Learning. Proceedings of the 2nd ACM Int. Conf. on AI in Finance (New York, U.S.A./Association for Computing Machiney) pp 1-8.
[12]
Yang H, Liu X Y, Zhong S. 2020. Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy. Proceedings of the 1st ACM Int. Conf. on AI in Finance (New York, U.S.A./Association for Computing Machiney) pp 1-8.

Index Terms

  1. A novel portfolio strategy approach using deep reinforcement learning

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM Other conferences
    ICMLCA '23: Proceedings of the 2023 4th International Conference on Machine Learning and Computer Application
    October 2023
    1065 pages
    ISBN:9798400709449
    DOI:10.1145/3650215
    Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 16 April 2024

    Permissions

    Request permissions for this article.

    Check for updates

    Qualifiers

    • Research-article
    • Research
    • Refereed limited

    Funding Sources

    Conference

    ICMLCA 2023

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • 0
      Total Citations
    • 13
      Total Downloads
    • Downloads (Last 12 months)13
    • Downloads (Last 6 weeks)1
    Reflects downloads up to 17 Feb 2025

    Other Metrics

    Citations

    View Options

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    HTML Format

    View this article in HTML Format.

    HTML Format

    Figures

    Tables

    Media

    Share

    Share

    Share this Publication link

    Share on social media