Estimating CO₂ emissions using a fractional grey Bernoulli model with time power term

Area of Science:

• Environmental Science
• Climate Change Modeling
• Econometrics

Background:

• Global warming, driven by carbon dioxide (CO2) emissions, poses significant threats to human health and quality of life.
• Accurate CO2 emission predictions are vital for policymakers to develop effective low-carbon environmental strategies.

Purpose of the Study:

• To introduce and validate a novel fractional grey Bernoulli model (FGBM(1,1,[Formula: see text])) for enhanced CO2 emission prediction.
• To assess the adaptability and effectiveness of the FGBM(1,1,[Formula: see text]) against existing grey prediction models.

Main Methods:

• Development of the FGBM(1,1,[Formula: see text]) model, highlighting its adaptable nature through parameter adjustments.
• Utilizing four carbon emission datasets, optimizing model parameters with three algorithms, and employing two evaluation criteria for accuracy assessment.
• Comparative analysis of FGBM(1,1,[Formula: see text]) against six other grey models to establish its superior feasibility and effectiveness.

Main Results:

• The FGBM(1,1,[Formula: see text]) model demonstrated higher accuracy and effectiveness compared to six other grey models.
• Optimized parameter selection and rigorous evaluation confirmed the model's predictive capabilities.

Conclusions:

• The FGBM(1,1,[Formula: see text]) model provides a feasible and effective tool for predicting CO2 emissions.
• Forecasts indicate a gradual rise in CO2 emissions for India, the Asia Pacific region, and globally from 2020-2024, while US emissions are projected to decline slightly.