Principal Component Analysis and Neural Network Ensemble Based Economic Forecasting

The application of neural network ensemble (NNE) to economic forecasting can heighten the generalization ability of learning systems through training multiple neural networks and combining their results. In this paper, principal component analysis (PCA) is developed to extract the principal component of the economic data under the prerequisite that the main information of original economic data is not lost, and the input nodes of forecasting model are effectively reduced. Based on Bagging, a NNE constituted by five BP neural networks is employed to forecast GDP of Jiangmen, Guangdong with favorable results obtained, which shows that NNE is superior to simplex neural network, and valid and feasible for economic forecasting.