Machine Learning-Based Modeling of the Environmental Degradation, Institutional Quality, and Economic Growth

摘要

Abstract This study was aimed at investigating the determinants of environmental sustainability in 86 countries from 2007 to 2018. The natural gradient boosting (NGBoost) algorithm was implemented along with five machine learning models to forecast the trends of CO2 emissions. In addition, the SHapley Additive exPlanation (SHAP) technique was used to interpret the findings and analyze the contribution of the individual factors. The empirical results indicated that the predictions obtained using NGBoost were more accurate than those obtained using other models. The SHAP value exhibited a positive correlation among the amount of CO2 emissions, economic growth, and opportunity entrepreneurship. A negative correlation was observed among the governance, personnel freedom, education, and pollution.