The Influence of Training Step on Price Forecasting Based on Back Propagation Neural Network

In order to obtain high stability and generic model in price forecasting, the training steps is used to improve forecasting model in this paper. The data of agricultural products and mobile products extracted from Website based on improved Back Propagation (BP) neural network are obtain the different results. Experiment results show the training steps affect the efficiency is limited. The different influence of TS base on two types of products, and we obtain the better length of training step between 3 with 5.