Abstract. This study investigates the impact of energy consumption, natural resource use, foreign direct investment (FDI), and trade openness on environmental degradation in Algeria over the period 1970–2022. To achieve this, we applied advanced econometric and machine learning approaches, including the Augmented ARDL model, the New DYARDL framework, and the Kernel Regularized Least Squares (KRLS) technique. The empirical results revealed divergent findings regarding the role of energy consumption: while traditional econometric models identified it as a primary driver of emissions, the dynamic specification attenuated its significance, reflecting the increasing reliance on relatively less polluting natural gas. Furthermore, the results provided evidence in support of the pollution haven hypothesis for trade openness and indicated that natural resource use contributes positively to environmental quality. In contrast, FDI exhibited no significant or consistent effect, as its scale remained insufficient to generate measurable environmental outcomes. Finally, the KRLS results confirmed the presence of complex and nonlinear interactions between economic and environmental factors.

Keywords: environmental quality, energy consumption, natural resources, trade openness and FDI, DYARDL and KRLS.