Ansari Saleh Ahmar, Zulkifli Rais, Faika Tunnas
This study compares two forecasting methods, Support Vector Regression (SVR) and SutteARIMA, to predict coal prices in Indonesia using monthly data from January 2013 to October 2022 provided by the Ministry of Energy and Mineral Resources. Uniquely, this study integrates the SVR approach with various kernel settings and finds that the radial setting provides the best results, demonstrating a high level of precision in predicting coal prices. However, the SutteARIMA method showed even greater accuracy, making it the more reliable choice between the two. The novelty of this study lies in the use of SutteARIMA, an adaptation of the traditional ARIMA method tailored to enhance accuracy in the context of coal price data in Indonesia. The findings suggest that the SutteARIMA method not only captures the trends in coal price movements with more precision but also provides forecasts that could be more useful for policymakers and stakeholders in the energy sector. By employing the more accurate SutteARIMA method, decision-makers can better anticipate price changes, which is crucial for economic planning and strategic resource management. This study highlights the importance of choosing the right forecasting tool to support policy development and economic decisions in Indonesia’s coal market. © The Author(s), under exclusive licence to Springer Nature B.V. 2025.
Department of Statistics, Faculty of Mathematics and Natural Science, Universitas Negeri Makassar, Makassar, Indonesia
