How mineral resources rent collaborate with consumer price index, environmental policies, and economic performance in Türkiye and India: Evidence from artificial neural networks and machine learning

Abstract

AbstractTaking focus on the possible effects on welfare and environmental issues in Türkiye and India, this study explores the relationship between the leasing of mineral resources (MRs), economic performance, use of renewable energy, and environmental policies. The study estimates changes in MRs throughout economic expansion using artificial intelligence (artificial neural network [ANN]) and supervised machine learning (SML). It focuses on important variables like index of stringency of environmental policies and the consumer price index, the conclusions of the ANN, ensemble method, and ML studies show how sensitive quarterly changes in the rent on MRs are to changes in the consumer price index, economic performance, and the use of renewable energy. Evaluation criteria such as root mean square error (RMSE), mean absolute error (MAE), mean square error (MSE), mean absolute percentage error (MAPE), and coefficient of determination highlight how much better ML models predict outcomes than ANN trials. In particular, the ML findings show an outstanding R2 of 0.99, an MAE of 0.6625, an MSE of 0.8324, a MAPE of 35.3677, and an RMSE of 0.9123 for India. Türkiye's machine learning results, on the other hand, display an MAE of 0.0164, an MSE of 0.0007, MAPE of 66.1594, RMSE of 0.0279, and a strong R2 of 0.98. For ANN, the error histogram is plotted to assess the model. The extremely low value of 0.0090 and 0.010, respectively, for Türkiye and India on the error histogram reflects the exceptional prediction quality. Türkiye and India have abundant MRs; however, they must be managed correctly for long‐term sustainability. Future researchers may verify this work using time series or panel data from other disciplines. This study examines factors affecting sustainable economic growth, including MR use, environmental policies, and eco‐friendly innovations. Other indicators, such as energy efficiency, carbon dioxide emissions, renewable energy consumption, and global value chain participation, may provide a different perspective. This study's conclusions should be verified by more research employing other geographic locations and others machine learning methods, as well as to illustrate how sustainable development is influenced by other variables.