Macroeconomic forecasting algorithm based on novel adaptive neural network

With respect to the existing problems of artificial neural network (ANN) methods for economic modeling and forecasting, an adaptive ANN algorithm is proposed after a brief introduction to the features of macroeconomic system. The training data set of the ANN was enlarged by using a chain style data recombination method first. Then the increasing rates of the input items and the so called time window data were employed as input layer items of the ANN. Simulation results tell that the proposed algorithm can improve the generalization ability and forecasting performance of the ANN model efficiently.