AUTHOR=Jiang Jianming , Ban Yandong , Nong Sheng 

TITLE=Novel disctete grey Bernoulli seasonal model with a time powter term for predicting monthly carbon dioxide emissions in the United States

JOURNAL=Frontiers in Environmental Science

VOLUME=Volume 12 - 2024

YEAR=2025

URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2024.1513387

DOI=10.3389/fenvs.2024.1513387

ISSN=2296-665X

ABSTRACT=This study proposes a more efficient discrete grey prediction model to describe the seasonalvariation trends of carbon dioxide emissions. The setting of the bernoulli parameter and the time powerterm parameter in the new model ensures that the model can capture the trend of nonlinear changesin the sequence. At the same time, the inclusion of dummy variables allows for the direct simulationof seasonal fluctuations in carbon dioxide emissions without the need for additional treatment of theseasonality in the sequence. The optimal search for the model’s hyperparameters is achieved using the MPA algorithm. The constructed model is applied to the monthly U.S. carbon dioxide emissions datafrom January 2003 to December 2022, a total of 240 months. The model is trained on 216 months of datafrom January 2003 to December 2020, and the monthly data from January 2021 to December 2022 is usedfor prediction, which is then compared with the actual values. The results show that the proposed modelexhibits higher forecasting performance compared to SARIMA and other models. Therefore, this methodcan effectively simulate the seasonal variation trends in carbon dioxide emissions, providing valuablereference information for relevant departments to formulate more effective policies.