Research On Domestic And Foreign Stock Market Forecast Based On BP Neural Network Stock Index Forecasting System

Internet financial and quantitative investment is increasingly important in the context of big data,a large number of stock analysis and stock price forecasting system was developed to practice.Due to the uncertainty of the structure of the stock market system and the complexity of the external economic environment factors,it is very difficult to forecast the stock market.This paper analyzes the application of various stock price forecasting methods in the stock market forecasting problem,and uses BP neural network to forecast the stock price.Neural network is a black box theory,at the same time with self-learning,adaptive and other characteristics,not affected by the reasons for data fluctuations do not need complex mathematical modeling.The neural network is very close to the nonlinear system,which can avoid the influence of human factors and expand the limitation of statistical quantitative forecasting method.The first chapter of this paper is the introduction,introduces the background and significance of the topic,the status quo of research at home and abroad,the research methods,ideas,article structure and innovation of this paper.The second chapter is an overview of stock market forecast and BP neural network.It introduces the basic knowledge of stocks,the characteristics of stock market and the forecasting methods of stock market both at home and abroad.The principle structure of neural network is introduced,the basic concepts,characteristics and BP neural network algorithm with the corresponding MATLAB toolbox program.The third chapter is the design of BP neural network system.Firstly,the system architecture design is carried out.Secondly,the system function is designed,including the system management module,the data management module and the stock index forecasting module.The detailed design of the core module of stock index prediction is carried out again.Finally,the database design is carried out.The fourth chapter is based on BP neural network stock forecasting system of domestic and foreign stock market forecast,this part through the establishment of the model,the original data pretreatment training neural network model to determine the model parameters after the use of BP neural network stock forecasting system for empirical analysis,(Stock code BABA),the NASDAQ Stock Exchange(stock code MSFT),the Shanghai Stock Exchange’s Sany Heavy Industry Stock Exchange(stock code 600031)and the Shenzhen Stock Exchange(stock code BABA),the New York Stock Exchange’s Alibaba stock(stock code BABA)Of the Ping An Bank stock(stock code 000001),a total of four stocks to carry out domestic and international stock price forecast empirical research,The results show that the BP neural network stock index forecasting system can effectively predict the stock price,while the accuracy of US stock price forecast is lower than the domestic stock price forecasting accuracy,which indicates that the mature market stock price is more difficult to predict,The Validity of Stock Price Forecasting Method for BP Neural Network Stock Index Forecasting System.The fifth chapter is the conclusion and the prospect,summarizes the forecasting ability of the stock index forecasting system in our country draws the conclusion and points out the shortcomings of this article and the future research direction.