Comparison of DQN And Double DQN Reinforcement Learning Algorithms for Stock Market Prediction

DOI

10.2991/978-94-6463-370-2_19How to use a DOI?

Keywords

Reinforcement Learning; Stock Prediction; Deep Q-Network; Double Deep Q-Network

Abstract

The financial industry has always considered stock market forecasting to be vital. In recent years, the application of reinforcement learning techniques in stock market prediction has gained attention. This study aims to explore using Deep Q-Networks (DQN) and Double Deep Q-Network (DDQN) for stock market prediction. Historical stock prices and relevant market data are used as inputs to construct a reinforcement learning environment for training the DQN and DDQN models. The objective of these models is to predict the future price trends of stocks by learning optimal policies. Results demonstrate that both DQN and DDQN models exhibit strong performance in stock market prediction tasks. They are able to capture the non-linear characteristics and dynamic changes of the stock market more accurately compared to traditional indicator-based methods. Furthermore, the DDQN model shows slightly superior results in certain metrics, indicating that the use of target networks for stable training can improve prediction performance. The findings hold significance for investors and financial institutions, providing valuable insights for investment strategies and risk management. Additionally, by exploring the application of reinforcement learning methods in stock market prediction, this research offers new perspectives for further studies in the financial domain. However, the complexity and uncertainty of the market may impact prediction performance. Future research can focus on enhancing model architectures, optimizing training algorithms, and considering the incorporation of additional market information to improve the accuracy and robustness of predictions.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Langxi Gao
PY  - 2024
DA  - 2024/02/14
TI  - Comparison of DQN And Double DQN Reinforcement Learning Algorithms for Stock Market Prediction
BT  - Proceedings of the 2023 International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2023)
PB  - Atlantis Press
SP  - 169
EP  - 177
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-370-2_19
DO  - 10.2991/978-94-6463-370-2_19
ID  - Gao2024
ER  -