A Gray Model With a Time Varying Weighted Generating Operator

A grey model with time varying weighted generating operator is put forward in order to fully extract information concealed in recent data. This model increases the weight of new data and relieve the influence of the the data fluctuation. The relationship between the sample size and the error from the inverse time varying weighted generating operator is discussed. Compared with traditional grey forecasting models, the results of practical numerical examples demonstrate that this grey model performs well in forecasting problems with limited data, and can provide reliable and acceptable accuracy for future prediction.