Shanghai Composite Pca-bp Model Predicted

As a high-risk and high return of investment, the problem of predicting stock market prices has been a hot issue of the financial system. In recent years, investors have been committed to finding effective forecasting methods and tools for stock market, but also reveals some of the internal operation of the stock market rules, however, Taking into account the complexity of the internal structure and the variability of external factors, the existing prediction result is'not satisfactory, the development of artificial neural network provides an effective way for solving this problem.As a massively parallel processing systems, Artificial neural networks build the model based on the internal relations, with good self-learning ability, and strong anti-interference ability, which has achieved satisfactory results in the short-term prediction of stock price,however,there are some weaknesses, On the basis of previous experience this paper propose to build model that using the method of combining the principal component analysis and the BP neural networks for building effective solution to this problem.First of all, the paper adopt statistical theory of principal component analysis and artificial neural network knowledge, using SPSS statistical analysis software and analyze the factors which affects the Shanghai Composite Index stocks from the macro environment to the External market and from the fundamentals to the technical based on the auxiliary of MATLAB language, then got the main factor which affect the Shanghai Composite Index; secondly, we build an artificial neural network model, using the selected set of training and testing samples to verify the accuracy of the model. Finally, we suggest some opinions about the use and amendment of the model, which make the model can be more effective and practical.