An Evidence-Based Explainable AI Approach for Analyzing the Influence of CO 2 $$_{2}$$ Emissions on Sustainable Economic Growth

Macroeconomic indicators play a crucial role in the development and overall sustainable economic growth of any country. This research focuses on analyzing time series data to explore the connection between CO 2 $$_{2}$$ emissions and GDP per capita. We addressed this challenge by developing a novel hybrid sequential model named the Multi-Recurrent Fusion (MRF) model. By incorporating the strength of GRU, LSTM, and Bi-LSTM models, the proposed MRF model surpassed other traditional deep learning models with an encouraging R 2 $$^{2}$$ score of 83.31%. Additionally, the minimal error rates denote the supremacy of MRF over other models utilized. This study aims to investigate the factors that affect sustainable economic growth, specifically focusing on the role of CO 2 $$_{2}$$ emissions using explainable AI tools like SHAP and ELI5. The findings offer valuable insights into the factors influencing macroeconomic trends and strongly argue that various emissions have no long-term relationship with income growth. This research demonstrates the potential of advanced AI techniques in enhancing our understanding of economic and environmental interactions, highlighting the incapability of traditional econometric models and challenging the previous results.