Abstract: This paper presents a method for stock market modeling and forecasting based on fuzzy neural network, genetic algorithm is used to learn the connection weights of the fuzzy neural network and partitions of fuzzy subsets. It has been shown by the modeling and forecasting results about Shanghai stock market price index and a company stock price that the method has reinforcement learning properties and mapping capabilities. With respect to modeling and forecasting of nonlinear system which has some uncertainties,the method is available.