Research on quantitative timing trading strategy based on deep reinforcement learning

Quantitative trading is an important research direction in the financial field. With the perfect development of the financial market, the trading strategies that want to obtain stable returns are becoming more and more complex, and many traditional trading strategies are gradually ineffective. In recent years, with the rapid development of artificial intelligence, the powerful data processing and analysis ability of deep learning algorithms makes it able to analyze and process the massive data generated by the financial market. This paper proposes a deep reinforcement learning algorithm, EDPG, which combines GRU and DDPG. It uses the ability of GRU sequence feature extraction and DDPG to analyze and make decisions to find the market operation law make rational trading decisions. Experiments on some Chinese A-share stocks have achieved good returns and exceeded the traditional quantitative strategy.