A modified grey forecasting model for long-term prediction

Grey theory is a truly multidisciplinary and generic theory that deals with systems which are characterized by poor information and/or for which information is lacking. In this paper, a modified grey model, combined with a simple statistical method to determine the model coefficient and a sectional model, by using another variable to modify the original grey prediction model for long-term forecasting, is proposed. This new method not only can improve the prediction accuracy of the original grey model, but also can make it suitable for long-term forecasting. Finally, we use power demand forecasting in Taiwan for our case study to test the efficiency and accuracy of the proposed method.