Abstract
Portfolio management involves position sizing and resource allocation. Traditional and generic portfolio strategies require forecasting of future stock prices as model inputs, which is not a trivial task since those values are difficult to obtain in the real-world applications. To overcome the above limitations and provide a better solution for portfolio management, we developed a Portfolio Management System (PMS) using reinforcement learning with two neural networks (CNN and RNN). A novel reward function involving Sharpe ratios is also proposed to evaluate the performance of the developed systems. Experimental results indicate that the PMS with the Sharpe ratio reward function exhibits outstanding performance, increasing return by 39.0% and decreasing drawdown by 13.7% on average compared to the reward function of trading return. In addition, the proposed PMS_CNN model is more suitable for the construction of a reinforcement learning portfolio, but has 1.98 times more drawdown risk than the PMS_RNN. Among the conducted datasets, the PMS outperforms the benchmark strategies in TW50 and traditional stocks, but is inferior to a benchmark strategy in the financial dataset. The PMS is profitable, effective, and offers lower investment risk among almost all datasets. The novel reward function involving the Sharpe ratio enhances performance, and well supports resource-allocation for empirical stock trading.
Highlights
Most conventional trading strategies generate trading signals based on predetermined subjective indicators, Common portfolio strategies, such as modern portfolio theory (MPT) [7] and the Kelly criterion [8], require predictions pertaining to future stock prices as inputs for portfolio management
There has been considerable research into the selection of commodities, position sizing, and resource allocation [5, 6]
In this paper, we sought to overcome the above-mentioned limitations by developing an efficient Portfolio Management System (PMS) through the implementation of conventional neural networks (CNN) and recurrent neural networks (RNN) networks within the Reinforcement learning (RL) architecture in order to support decision-making in the allocation of resources
Summary
Common portfolio strategies, such as modern portfolio theory (MPT) [7] and the Kelly criterion [8], require predictions pertaining to future stock prices as inputs for portfolio management. In this paper, we sought to overcome the above-mentioned limitations by developing an efficient Portfolio Management System (PMS) through the implementation of CNN and RNN networks within the RL architecture in order to support decision-making in the allocation of resources. We use two NN-based models (CNN and RNN) to deal with spatial and temporal information in order to refine the portfolio strategy This makes it possible for the PMS CNN and PMS RNN systems to assign appropriate weights to stocks to assist in the allocation of resources for each training day. The proposed PMS remains profitable and low-risk regardless of the dataset, and the novel reward functions involving the Sharpe ratio and return factors truly enhance performance. Proposed a RL-based portfolio management system, concatenated with CNN and RNN networks to support resource-allocation for empirical stock trading. Experiment results demonstrate the applicability of CNN to the formulation of an EMN portfolio as well as the scalability of the proposed PMS to a variety of datasets
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
