The Improvement Of Back Propagation Algorithm

In recent years, the research about artificial neural network have gotten important development. It inspires many scientists' enthusiasm and interest in the field of computer science , brain neural science and artificial intelligence. Artificial neural network is simulation to the information-processing mechanism of the brain. It is expected that realize the brain function by simulating the structure and thinking of brain. Its theory have been applied in many fields.The multilayer perception, trained by the back propagation algorithm , is currently the most widely used neural network. It solve the question of hidden layer learning rule that utilize the method of error back propagation . The essence of back propagation networks is that make the change of weights become little by gradient descent method and finally attain the minimal error. Since it adopt the steepest descent method in nonlinear programming, it has the drawback that easy converge to a local minimum point of the error function.In this paper, we improve the objective function of back propagation algorithm based on different financial actual situation. Change the objective function into the expectation and the variance of error function to make its application more wide. And we put forward a new algorithm that can solve the drawback of back propagation that easily converge to a local minimum point of the error function. Then we proved the convergence of this algorithm by using the characters of Markov chain and the effectiveness of this algorithm by example. Finally we try to forecast the stock price by BP network and find that the lag exist. So it isn't fit for stock price forecasting by on-line way through the test of several stocks in Shanghai stock exchange.