Application of Innovations Feedback Neural Networks in the Prediction of Ups and Downs Value of Stock Market

摘要

This paper presents how the idea of Kalman prediction can be used to the study of the prediction of neural networks and puts forward an Innovations Feedback Neural Networks (IFNN). IFNN adopts Back Propagation (BP) algorithm and is applied to the prediction of ups and downs value of the stock. The theoretical results and numerical experiment show that the predicted capability of IFNN is better than that of the normal feedforward networks. Finally, a relative completed stock prediction system is developed on Windows 2003/XP platform by OOP programming method and sql server2000 database. By practical operation in Resource Investment Consult Inc., it shows the system can improve prediction credibility of stock market effectively.