Modelling CO2 Emissions Through Econometric and Machine Learning Approaches: The Role of Renewable Energy and Resource Use in Climate Action

This research delves into the complex relationships among carbon dioxide (CO2) emissions, resource use, and energy consumption, showing how renewable energy adoption and reliance on fossil fuels shape greenhouse gas emissions. Leveraging panel data from 120 countries over a 10-year span, the analysis combines econometric approaches, including Weighted Least Squares (WLS), Generalised Linear Models (GLM) and Generalised Additive Models (GAM), with machine learning techniques such as Multivariate Adaptive Regression Splines (MARS) to capture nonlinear dynamics. Robustness checks are conducted using bagging, boosting, and Support Vector Machines (SVM) algorithms. Our analysis reveals strong associations between CO2 emissions and key factors such as fossil fuel energy consumption, methane emissions, and renewable energy use. These results underscore the critical need for a global shift toward renewable energy systems, enhanced resource management, and well-targeted policy interventions. By employing cutting-edge methodologies, this study suggests harmonising economic growth with sustainability objectives and advancing climate change mitigation efforts.