The impact of socio-economic factors on the ecological footprint in Turkey: A comprehensive analysis using machine learning approaches

This study undertakes a comprehensive investigation into the impact of socioeconomic factors on the ecological footprint (EFP) in Turkey. It employs robust econometric methods and advanced Machine Learning (ML) models. The study employs Generalized Linear Models (GLM), Autoregressive Integrated Moving Average (ARIMA) models, Robust Least Squares (RLS) regression, and Granger causality tests to identify electric power consumption, real GDP, and life expectancy as significant positive drivers of the EFP. At the same time, trade openness and urbanization negatively impact the dependent variable. Advanced machine learning models, including Artificial Neural Networks (ANN) and Recurrent Neural Networks (RNN) such as Long Short-Term Memory (LSTM) models, corroborate these findings, enhancing the study's comprehensiveness. The results underscore the importance of promoting energy efficiency, green growth, sustainable trade practices, and resource-efficient urban development to mitigate environmental impacts. The study provides robust empirical evidence and policy recommendations for reducing the EFP in Turkey, emphasizing the integration of sustainable practices in socioeconomic activities.