The optimal combination forecasting model based on closeness degree and IOWHA operator under the uncertain environment

摘要

We propose the optimal combination forecasting model based on closeness degree and induced ordered weighted harmonic averaging (IOWHA) operator under the uncertain environment in which the raw data are expressed as interval numbers. It is a new kind of combination forecasting model with variant weights. We can obtain weighted coefficient vectors of combination forecasting methods by maximizing the closeness degree of combination forecasting, instead of minimizing absolute errors. We put forward the new concepts of closeness degree for the center and radius of interval numbers sequences, and construct the optimal interval combination forecasting model to maximize the sum of convex combination with closeness degree of interval center and closeness degree of interval radius. Furthermore, we discuss the solution to the model. Finally, an example is used to show that this model can improve the combination forecasting accuracy efficiently compared with each individual forecasting method.