Stock Market Forecasting Algorithm Based on Improved Neural Network

This paper further analyzes the problems in stock short market forecasting and compares multiple stock price forecasting algorithms. It also discusses the feasibility of BP neural network, PCA method and Genetic Algorithm in short market forecasting. For the defects of traditional BP algorithm which often traps into local minima, in forecasting accuracy, we optimize the BP neural network and establish a GA-BP algorithm based forecasting model. The experiments adopt the Shanghai index data to make simulation and provide corresponding error analysis. The results show that the GABP model proposed in this paper has certain improvement in stock price forecasting accuracy.