The Research Of The Adjustment Of Stock Prices To New Information Based On RBF Neural Networks

At present, the main methods for the prediction of the stock price ARE fundamental analysis and technical analysis. However, the stock price is influenced by variety factors, experts, scholars, as well as institutional and individual investors are studying the stock price changes from different angles, using different theories and methods, and expect to make a more accurate prediction of stock price. In the course of the study, a lot of theories and prediction method were carried out based on different theoretical foundation and academic point of view.Fundamental analysis is the study of the status of the company, the development trend of the industry fundamentals to estimate the fair value of the company, and believe that the company's share price in the long term reflecting the fair value. While technical analysis is predicting the stock price with a variety of historical technical indicators. According to the fundamental theory, the announcement of new information will change the investors' view on the company's fundamental, and thus have an impact on the stock price. There are a large number of studies have confirmed that the Chinese stock market is in weak-efficiency, which means the new information will not only affect the stock price, and this influence will last for a period of time. This paper uses the RBF neural network as a tool, combined with the method of fundamental analysis and technical analysis, to study the new information's impact on the stock price.According to the study, whether the company announces the annual report, result notice or dividend plan, all of that information will have an impact on the price of the stock. And this kind of influence is often emerged in the few trading days before the announcement date, and continues to several trading day after the announcement.