Research on Quantitative Trading Based on Deep Learning

Traditional quantitative trading strategies are widely used in stocks, futures and other financial markets, but the manual extraction method of features makes it lack the ability to effectively adjust strategies dynamically, and deep reinforcement learning can effectively simulate complex market environments and solve dynamic quantitative trading problems. Based on the development status of the financial industry, this paper introduces the deep reinforcement learning algorithm into the field of stock trading to build an intelligent trading model. Its goal is to discover the laws of the market in the learning of massive data, so as to carry out effective transactions, effectively avoid market risks and improve investors’ returns. On the basis of the traditional DQN algorithm, corresponding to the actual requirements, we propose RB_DRL deep reinforcement learning algorithm model to improve the network structure. The experimental analysis results show that the improved model also shows good results in multi-group comparative experiments.