Utilizing the Group Method of Data Handling (GMDH) for Electric Demand Forecasting: A Case Study of Turkey

This paper presents a comprehensive load forecasting study focusing on Turkey, utilizing weekly, monthly, and yearly data. The investigation compares two prominent forecasting methodologies: the Group Method of Data Handling (GMDH) and the Adaptive Neuro-Fuzzy Inference System (AN-FIS). Through empirical analysis, it is demonstrated that GMDH outperforms ANFIS in terms of accuracy and reliability for load forecasting in this specific context. This research contributes valuable insights into enhancing forecasting models for energy demand prediction, particularly in the dynamic and evolving energy landscape of Turkey.