The Research Of Stock Price Forecast Basing On Wavelet Analysis And Neural Network

Stock market became the most important and absolutely necessary part of securities trade and financial market after the establishment in China. More and more investor pay attention to stock market.So the analysis and forecast of stock market have not only signification of theory but also merit of application.As the important part of stock market-Technical Analysis has achieved great progress in USA and Europe.New theory and technology has breathed into analysis of Stock market along with the development of information technology.The demand of new analysis technique of Chinese stock market is an impetus of this study. In this paper, the author introduce the general situation and research in existence of Chinese stock market firstly.Then the elements and development of Artficial Neural Network.Finally,the principle and applications of Wavelet is presented.After the intrudction,the author discuss the possibility of the combine of Artficial Neural Network and Wavelet analysis and its application in research of Chinese Stock market.The author recommend a new way that searching the trend of stock price by Wavelet firstly,training the Artficial Neural Network secondly.The new way is help to solve the problem such as inferior performance?failure of convergence?error in trend. To test the validity of the new analysis means,more than 30 stocks was selected and some experiments were done.The results of the experiments show that the right ratio of the forecast reaches 70%.We have enough reason to believe that the new method is better than the technology in existence. In the last part,the author introduce the a new analysis system that combine the ability of C++Builder and calculate capability of Matlab.The core of the system is that provide the friendly interface by using C++Builder and prediction of price by using Matlab.