The securities market is a highly complex nonlinear dynamic system, its changes of time sequence related to impacts of political,economic,psychological and many other uncertain factors. In order to make accurate forecasts to the price of securities, we need a prediction method systematically response to the price of securities under the common function of many factors. On the highly non-linear characteristic of the securities market, this paper designs a genetic algorithm optimization BP neural networks model to forecast the price of the securities.This paper connects genetic algorithm with BP neural networks to establish a genetic BP neural networks. On the genetic algorithm parameters and BP neural networks parameters settings carries out a detailed study, describes the genetic BP algorithm steps. Use of the three-tier genetic BP neural networks establishes forecast model for the stock market of the financial securities market .Through stock forecast testing, certificates genetic BP neural networks model has a good prediction for securities prices. |