Development of a Multi-Objective Optimisation Framework for Risk-Aware Fractional Investment Using Reinforcement Learning in Retail Finance
DOI:
https://doi.org/10.5281/zenodo.16667023Keywords:
Reinforcement learning, fractional investment, retail finance, multi-objective optimisation, portfolio management, risk-aware strategiesAbstract
Retail investors often face a complex trade-off between maximising returns, minimising risk, and achieving portfolio diversification, especially in the context of fractional investment. This study proposes a novel multi-objective optimisation framework employing reinforcement learning (RL) to address these competing goals in retail finance. The framework integrates a risk-aware reward function with dynamic portfolio rebalancing strategies to allow fractional investment across multiple assets. By utilising deep Q-learning and Proximal Policy Optimisation (PPO), the system learns optimal policies under varying market conditions, incorporating investor risk preferences and transaction constraints. Empirical testing was conducted on historical financial data spanning equities, ETFs, and cryptocurrencies. The results demonstrate superior portfolio performance in terms of Sharpe ratio, maximum drawdown, and diversification entropy when compared to traditional portfolio allocation techniques. Our framework paves the way for democratised, intelligent retail investment platforms that are both adaptive and aligned with user-specific financial goals.References
Nevmyvaka, Y., Feng, Y., & Kearns, M. (2006). Reinforcement learning for optimized trade execution. In Proceedings of the 23rd International Conference on Machine Learning (pp. 673–680). ACM. https://doi.org/10.1145/1143844.1143929
Morimura, T., Sugiyama, M., Kashima, H., Hachiya, H., & Tanaka, T. (2010). Nonparametric return distribution approximation for reinforcement learning. In Proceedings of the 27th International Conference on Machine Learning(pp. 799–806). Omnipress. https://doi.org/10.5555/3104322.3104424
Chow, Y., Tamar, A., Mannor, S., & Pavone, M. (2015). Risk-sensitive and robust decision-making: A CVaR optimization approach. In Advances in Neural Information Processing Systems (Vol. 28, pp. 1522–1530). Curran
Jiang, Z., Xu, D., & Liang, J. (2017). A deep reinforcement learning framework for the financial portfolio management problem (arXiv:1706.10059). arXiv.
Xiong, Z., Liu, X. Y., Zhong, S., Walid, A., & Yu, F. R. (2018). Practical deep reinforcement learning approach for stock trading (arXiv:1811.07522). arXiv.
Liang, Z., Chen, W., Zhu, Y., Jiang, J., Li, Y., & Li, J. (2018). Adversarial deep reinforcement learning in portfolio management (arXiv:1808.09940)
Wang, H., Zhang, Y., Li, K., & Wang, Y. (2020). Adaptive portfolio management via proximal policy optimization. Applied Soft Computing, 97, 106753. https://doi.org/10.1016/j.asoc.2020.106753
Tamar, A., Chow, Y., Ghavamzadeh, M., & Mannor, S. (2015). Policy gradient for coherent risk measures. In Advances in Neural Information Processing Systems (Vol. 28, pp. 1468–1476). Curran
Yang, S. (2023). Deep reinforcement learning for portfolio management. Knowledge-Based Systems, 278, 110905. https://doi.org/10.1016/j.knosys.2023.110905
Moody, J., & Saffell, M. (2001). Learning to trade via direct reinforcement. IEEE Transactions on Neural Networks, 12(4), 875–889. https://doi.org/10.1109/72.935086
Geng, W., Xiao, B., Li, R., Wei, N., Zhao, Z., & Zhang, H. (2023). Decomposition-based multi-agent distributional reinforcement learning for task-oriented UAV collaboration with noisy rewards. In 2023 International Conference on Wireless Communications and Signal Processing (WCSP) (pp. 342–347). IEEE. https://doi.org/10.1109/WCSP56871.2023.10155709
Almahdi, S., & Yang, S. Y. (2017). An adaptive portfolio trading system: A risk-return portfolio optimization using recurrent reinforcement learning with expected maximum drawdown. Expert Systems with Applications, 87, 267–279. https://doi.org/10.1016/j.eswa.2017.06.023
Deng, Y., Bao, F., Kong, Y., Ren, Z., & Dai, Q. (2016). Deep direct reinforcement learning for financial signal representation and trading. IEEE Transactions on Neural Networks and Learning Systems, 28(3), 653–664. https://doi.org/10.1109/TNNLS.2016.2522401
Bertsimas, D., & Pachamanova, D. (2008). Robust multiperiod portfolio management in the presence of transaction costs. Computational Management Science, 5(3), 163–195. https://doi.org/10.1007/s10287-007-0054-3
Wang, L., & Wang, X. (2024). An adaptive financial trading strategy based on proximal policy optimization and financial signal representation. Engineering Applications of Artificial Intelligence, 138(Part A), 109365. https://doi.org/10.1016/j.engappai.2023.109365
Park, H., Kim, J., Kim, J., Kang, Y., & Lee, K. (2020). An intelligent financial portfolio trading strategy using deep Q-learning. Expert Systems with Applications, 158, 113573. https://doi.org/10.1016/j.eswa.2020.113573
Ye, Y., Pei, H., Wang, B., Chen, P. Y., Zhu, Y., Xiao, J., & Li, B. (2020). Reinforcement-learning-based portfolio management with augmented asset-movement prediction states. In Proceedings of the 34th AAAI Conference on Artificial Intelligence (pp. 1110–1117)
Wang, R., Wei, H., An, B., Feng, Z., & Yao, J. (2020). Deep stock trading: A hierarchical reinforcement learning framework for portfolio optimization and order execution. In Proceedings of the 29th International Joint Conference on Artificial Intelligence (pp. 3965–3971)
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Jessy Christadoss, Manas Ranjan Panda, Bhakta Vaschal Samal, Girish Wali
This work is licensed under a Creative Commons Attribution 4.0 International License.