Multivariate Forecasting of Seasonal Carbon Dioxide Emissions via a Discrete Grey Multivariate Forecasting Model with a New Information Priority Accumulation Operator.

In this study, a more efficient Discrete Grey Multivariate Forecasting Model with A New Information Priority Accumulation Operation is proposed to depict the development trend of energy-related seasonal carbon dioxide emissions. The new information priority accumulation operation and an adaptive grey action quantity in the new model ensure excellent nonlinear fitting capabilities. The presence of the virtual variable allows the model to directly simulate seasonal fluctuations in seasonal carbon dioxide emissions without removing seasonal effects, showcasing the model's superiority. Therefore, the model can fit the nonlinear seasonal time series better. Experiments based on quarterly carbon dioxide emissions from energy consumption in the United States demonstrate the new method's optimal forecasting performance. Additionally, the optimization capability of each component in the new model is further validated by a more in-depth experiment. The effectiveness of this method in fitting seasonal carbon dioxide emissions is confirmed. [ABSTRACT FROM AUTHOR]