A novel weighted fractional TDGM model and quantum particle swarm optimization algorithm for carbon dioxide emissions forecasting

Abstract

This paper will present a novel weighted fractional TDGM(1,1) (WFTDGM) model based on the combination of the weighted fractional-accumulation generating operator (FAGO) and the TDGM(1,1) model. The suggested WFTDGM model would be able to reduce the traditional TDGM(1,1) model and the fractional TDGM(1,1), or FTDGM model when the parameters are adjusted differently. Hence, the quantum particle swarm optimization algorithm will be used to select the optimal parameters for the proposed model to achieve the best accuracy precision. Whereas the least squares estimate method is used to determine the remaining model parameters. Twenty numerical samples selected from various countries presented in this paper will be used as the case study. When compared to the other conventional grey models such as the GM(1,1), FGM(1,1), TDGM, and FTDGM models, the computational results indicate that the proposed model has the best forecast performance compared to the other models.