Application of the chain style data recombination methodand neural networks in macroeconomic forecast
链式数据重组与神经网络在经济预测中的应用

It is necessary to model predict and model the economic system and for evalusating economic circulation,simulation,and making development programs.Based on the characteristics of the macroeconomic forecast,such as small data sets and time_varying,a chain_style data recombination method is presented,which is used for the back_propagation (BP) neural networks training.This method was applied to real macroeconomic forecast,and achieved the higher forecasting precision than the original data sets.It also solves the problem how to decide the hidden node numbers for the BP neural networks.