Forecasting the CO2 Emissions at Malaysia using Group Method of Data Handling

Abstract

This study utilizes the Group Method of Data Handling (GMDH) model to anticipate CO2 emissions, providing a comprehensive analysis of various input variables, training methods, and forecasting stages. The research uncovers distinct patterns in the behavior of input variables X1, X2, and X3 during both training and forecasting. Notably, X2 consistently exhibits strong performance, whereas X1 and X3 face difficulties, particularly in the forecasting phase. The GMDH model demonstrates proficiency in adaptive self-organization, automatic feature extraction, managing non-linear relationships, and interpretability, enhancing its effectiveness in capturing complex patterns in CO2 emission data. The observed decrease in performance for specific inputs during the forecasting process highlights the necessity for improving and adjusting the model and developing a detailed grasp of the dynamics of the variables involved.

Notwithstanding these challenges, the study acknowledges the importance of CO2 emission forecasting for policymakers, industries, and environmentalists. It empowers them to make well-informed choices, implement measures to reduce harm, contribute to worldwide efforts to combat climate change and achieve emission reduction objectives. In essence, our research promotes a collective dedication to a future that is both environmentally friendly and resilient in the face of challenges.