Abstract

This paper selects the CSI 300 index as the research object, because its constituent stocks have strict selection criteria, the index has a high market coverage and uniform industry distribution, so it can basically reflect the stock market volatility. A total of 3800 sets of data from January 2006 to December 2021 are selected, and a training group and a test group are set up. Based on the research of BP neural network, PCA analysis method and GA genetic algorithm are cited first, and the advantages and disadvantages of each research method are introduced respectively. On this basis, a multi-combination prediction model based on BP neural network model is constructed, and the fitness function of GA genetic algorithm is improved, and finally the prediction performance of different models is derived through experiments. The experimental results show that the prediction results of the multi-combination prediction model are all better than the single BP neural network prediction model; the prediction results obtained by the improved GA-BP neural network model have 4% higher accuracy compared with the unimproved GA genetic algorithm; the efficiency of optimizing the number of variables is improved by about 55%.