Improving the accuracy of nonlinear combined forecasting using neural networks

The aim of the work presented in this paper is to propose artificial neural networks (ANN) as a tool for nonlinear combination of forecasts. In this s tudy, the performance of the networks is evaluated by comparing them to thr ee individual forecasting methods and two conventional linear combining met hods. The outcome of the comparison proved that the prediction by the ANN m ethod generally performs better than those by individual forecasting method s, as well as linear combining methods. The paper suggests that the ANN met hod can be used as an alternative to conventional linear combining methods to achieve greater forecasting accuracy. Meanwhile, ANNs also can be integr ated with many other approaches including connectionist expert systems to i mprove the prediction quality further. (C) 1999 Published by Elsevier Scien ce Ltd. All rights reserved.