Font Size: a A A

Stock Prediction Based On Neural Network Transaction System Design And Implementation

Posted on:2015-10-03Degree:MasterType:Thesis
Country:ChinaCandidate:T WuFull Text:PDF
GTID:2308330473952733Subject:Software engineering
Abstract/Summary:PDF Full Text Request
The stock as a financial market the most active, the most important financial assets and long-term investment instruments, while the stock prices as a barometer of pharos, stock market economy "". The stock market is a highly complicated nonlinear dynamic system, stock prices involves many uncertain factors, the influencing factors of the stock market is a very complex. To forecast stock price reasonable and effective has been the important issue of the financial field, is also a difficult problem. Example shows that many history, whether it is used in more formal esoteric financial economic theory, or for ordinary consumers, the price of the stock movement theory, price changes are all the people are most concerned about. Stock market, with the appropriate forecasting method to predict the stock price changes seriously analysis can reduce the risk of consumers to obtain financial securities industry, investment benefit.Methods firstly, can use the stock theory and BP neural network to predict the stock market, in the aspect of algorithm is improved compared to the other methods are advanced, some second, capable of using the basic method of software engineering is a complete system design, and it can run, has a certain practicality. Overall, stock forecasting system is successful. But at the same time there are some shortcomings, so it is the direction of future efforts. Only the algorithm constantly improve, real data stock forecasting system can meet the actual constantly in life, only the system constantly upgrading in order to run more perfect.In the design based on BP neural network prediction module, at the same time for the design of the stock trading system. The stock transaction process transactions in the accurate prediction, one is for testing the model to predict the stock, on the other hand, is to improve the overall system. The design of a complete stock forecasting, trading system...
Keywords/Search Tags:BP neural network, stock forecasting system, genetic algorithm, software engineering
PDF Full Text Request
Related items