Macroeconomic policies play a very important role in the development of the stock market. The impact of macroeconomic policies on the stock market is complex and nonlinear, and it is difficult for existing models to accurately predict.In order to improve the level of investment decision-making, this paper uses the deep deterministic policy gradient (DDPG) algorithm to study its application in predicting the impact of macroeconomic policies on the stock market. Through the collection of macroeconomic policies and historical stock data, an intensive learning model is established to predict changes in the stock market based on macroeconomic policies as environmental variables. After training the DDPG algorithm, the model learns the influence mechanism. The experiment analyzed from the three dimensions of volatility, prediction accuracy and return on investment. Compared with the SVM (Supported Vector Machine) and RF (Random Forest) algorithms, the average accuracy rate of the DDPG algorithm was 7.8% and 9.6% higher. Therefore, the DDPG algorithm can more accurately understand and grasp the impact of macroeconomic policies on the stock market, and effectively improve the level of investment decision-making and the rate of return. This article conclusion is of great significance to guide investors to make rational stock investment, and it also contributes to the healthy and stable development of the stock market.
