Application Research of Combined Forecasting Based on Induced Ordered Weighted Averaging Operator

Abstract

Aiming at the shortcomings of traditional weighted arithmetic combination forecasting model, using the induced ordered weighted averaging operator, according to the fitting accuracy of each single forecasting method at each time point of the sample interval to endue weighted, and with the error sum of squares as criterion, we establish a new combination forecasting model, which effectively improves the precision of combination forecasting. And the resident consumption levels were predicted using the model analysis.