Research on the Estimating Model of the Stock Market Price Based on the LM-BP Neural Network

Standard BP neural network is a most representative algorithm in the neural network model. But shortcomings exist in its process of application. For example: it´s hard to reach global optima, but can easily form local minimum, Low study efficiency and slow convergence rate appear because of the excessive training, the selection of the hidden layer nodes lack of theoretical guidance, in training, there is a tendency of forgetting the old samples while learning the new ones. The Levenberg-Marquardt algorithm refers to an optimization algorithm aiming at the global, which´s very suitable for neural network training. This paper has built an estimating model of the stock market price, based on the LM-BP neural network, and carries out a prediction on the stock market price. Also this paper has made a comparison of the prediction results with the standard BP neural network model. And this has reached a profound estimating effect.