Predicting the future course of energy consumption in line with sustainable development goals plays a critical role in long-term planning processes in terms of resource management, environmental sustainability, and economic stability. Therefore, in this study, Turkey's energy consumption until 2035 was estimated using the Random Forest machine learning model developed based on macroeconomic indicators. In the modeling process, energy consumption, economic growth, inflation, industrial carbon dioxide emissions and population growth data between 1965 and 2024 were used. These data were obtained from the World Development Indicators (WDI, 2025) and World Energy Statistics (WES, 2025) databases. In the analysis, the Random Forest model, a machine learning method, was preferred. In the established model, energy consumption was estimated to account for the nonlinear and complex interactions of economic growth, inflation, industrial carbon dioxide emissions and population growth variables. The estimation results obtained with the Random Forest model show that Turkey's energy consumption in the 2025-2034 period will follow a fluctuating but generally upward trend. Non-linear transitions have been observed in the forecasts, especially in some years, where there have been sudden jumps and fixations.
Primary Language | English |
---|---|
Subjects | Macroeconomic Theory |
Journal Section | Makaleler |
Authors | |
Early Pub Date | October 17, 2025 |
Publication Date | October 21, 2025 |
Submission Date | August 20, 2025 |
Acceptance Date | October 5, 2025 |
Published in Issue | Year 2025 Volume: 9 Issue: 2 |
Journal of Research in Economics is licensed under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
JORE is indexed in