A Novel Robust Grey Model for Forecasting Chinese Electricity Demand

Abstract

The grey model, which is abbreviated as GM (1, 1), has been widely applied in the fields of decision and prediction, particularly in the prediction of time series with few observations, referred to as the poor information and small sample in the literature related to grey model. Previous studies focus on improving the accuracy of prediction but pay less attention to the robustness of the grey model to outliers, which often occur in practice due to an incorrect record by chance or an accidental failure in equipment. To fill that void, we develop a robust grey model, whose structural parameters are obtained from the least trim squares, to forecast Chinese electricity demand. Also, we use the last value in the first-order accumulative generating time series as the initial value, according to the new information priority criterion. We name the novel grey model, proposed in this paper, the novel robust grey model integrating the new information priority criterion, which could be abbreviated as NIPC-GM (1, 1). In addition, we introduce a novel approach, that is, the bootstrapping test, to investigate the robustness against outliers for the novel robust grey model and the classical grey model, respectively. Using the data on Chinese electricity demand from 2011 to 2021, we find that not only does the novel robust grey model integrating the new information priority criterion have a better robustness to outliers than the classical grey model, but it also has a higher accuracy of prediction than the classical grey model. Finally, we apply the novel robust grey model integrating the new information priority criterion to forecasting the future values in Chinese electricity demand during the period 2022 to 2025. We see that Chinese electricity demand would continue to rise in the next four years.