The Research Of BP Neural Network In Stock Market Forecast

The analysis and prediction of the stock market is an eternal topic.There are great difficulties in its forecast because of the existence of system risk and non-system risk in stock market.There are many theoretical models for analysis and prediction of the stock market after so many years of development.But the Chinese stock market are different with the foreign financial market.On the one hand,its development is short;on the other hand,its environment has particularity and complexity.So many models abroad are not completely applicable to domestic stock market.As a rsult,there is a great trouble for forecasting the Chinese stock market.In this paper,we hope to produce a more ideal fitting prediction model by using the artificial neural network model in the background of the big data and artificial intelligence.The backpropagation nureal network is widely used.In the paper,the traditional time series analysis model and artificial neural network model based on the Shanghai stock market data were simulated.After that,we compare their performance of prediction and consider the effects of several important factors including model structure and training function.