Machine learning methods for investigating the nexus between FDI and environmental degradation: evidence from Sub-Saharan African countries

What the paper says

Purpose The purpose of the study is to investigate whether foreign direct investment (FDI) has an impact on environmental degradation in Sub-Saharan African (SSA) countries using machine learning (ML) methods for the years between 2002 and 2021. Design/methodology/approach In this study, k-nearest neighbour (k-NN), support vector machine (SVM) and random forest (RF) machine learning algorithms were used, and their performances were compared based on root mean squared error (RMSE), R-squared (R2), mean absolute error (MAE) criteria, respectively. While carbon dioxide (CO2) emissions are used as an indicator of environmental degradation, in addition to FDI, gross domestic product (GDP) per capita, urbanization, renewable energy consumption, trade openness, population density, natural resources rents, governance and inflation are used as explanatory variables in the study to obtain a comprehensive perspective. The dataset consists of 46 countries and is compiled from the World Development Indicators Database, World Bank. Findings Among ML methods, it is found that RF has the best performance based on the performance evaluation criteria. It is found that there is no evidence that FDI has an impact on environmental degradation in SSA countries and GDP per capita, renewable energy consumption and urbanization are the most important features affecting carbon dioxide emissions. According to the findings, renewable energy consumption positively affects environmental degradation, whereas GDP per capita and urbanization negatively affect. Originality/value In the literature, while the relationship between environmental degradation and FDI is examined through econometric analyses, there are very few studies using machine learning method to investigate this relationship. Therefore, this study is an attempt to fill this gap in the literature, and it provides valuable insights into the applicability of ML methods.