ARIMA component-switching models for forecasting CO2 transport emissions in the Middle East

Abstract This study proposes an innovative auto-regressive integrated moving average (ARIMA) component-switching model framework for forecasting carbon dioxide (CO2) transport emissions in the Middle East, targeting 11 countries across the region. Recognizing the critical role of accurate emissions forecasts for climate action, this research applies eight model variations to optimize predictive accuracy across diverse national contexts. The ARIMA (1,1,1) model emerged as particularly effective for Bahrain and Jordan, exhibiting strong predictive performance metrics, with low Bayesian information criterion and root mean square error values, confirming the model’s robustness. The analysis reveals a complex regional trajectory for CO2 emissions, with Iran, Iraq, Kuwait, Lebanon and Bahrain showing upward trends projected to intensify between 2025 and 2050. Notably, Kuwait, Lebanon and Iraq exhibit the highest expected growth, with emissions forecasted to more than double by the mid-21st century, highlighting these nations as key focus areas for intervention. Conversely, Saudi Arabia, the UAE, Qatar, Jordan, Syria and Yemen demonstrate promising downward trends, reflecting the potential impact of existing mitigation policies. This paper contributes valuable insights for policymakers and environmental planners by offering a forecasting model that not only accounts for historical data but also adapts to changing emission patterns. The ARIMA component-switching approach allows for tailored predictions across various nations, assisting in the strategic allocation of resources and the formulation of targeted policies to curb transport emissions. These findings support the broader goal of sustainable development and align with global climate commitments, offering a practical tool for emission reduction planning across the Middle East.