Probabilistic forecasting of short-term electricity prices in the Turkish day-ahead market

Year 2026, Volume: 10 Issue: 1 , 21 - 27 , 20.04.2026

Kübra Nur Akpınar

https://doi.org/10.35860/iarej.1820591

https://izlik.org/JA66HY27PH

Abstract

In modern power systems with increasing renewable energy integration, electricity price forecasting has become increasingly vital for system planning. This study focuses on Türkiye’s Day-Ahead Market (DAM) prices by utilizing a probabilistic machine learning model to improve short-term price prediction. A Quantile Gradient Boosting Regressor (GBR) was trained using hourly data obtained from the EPİAŞ transparency platform covering the period between 2022 and 2025. By estimating market-clearing prices, the model allows for capturing both the central tendency and the uncertainty of prices.
The model includes time stamp data as hour and day, as well as electricity generation resources and past prices. Quantitatively, the model achieved an RMSE of 434.82 TRY/MWh, a CRPS of 194.98 TRY/MWh, and a PICP of 0.74 for the 80% prediction interval. The results show that the proposed approach provides high-performance prediction intervals when compared with traditional single-point models. This probabilistic model could be used for decision-making in energy markets, as well as for the scheduling of renewable integrated storage systems within renewable energy systems.

Keywords

Day-Ahead Market , Gradient Boosting Regressor , Price Forecast , Probabilistic machine learning , Quantile regression

References

• 1. Özgüner, E., O.B. Tör, A.N. Güven, Probabilistic day-ahead system marginal price forecasting with ANN for the Turkish electricity market, Turkish Journal of Electrical Engineering and Computer Sciences. 2017. 25(6): p. 4923–4935.
• 2. Özcan, A.V. M. Erel-Özçevik, B. Karaman, İ. Baştürk, E. Zeydan, S. Taşkın, Ü. Çetinkaya, Forecasting day-ahead electricity prices for the electricity market with dynamic time period. Energy, 2025. 338: 138766.
• 3. Xia, S., G. Wang, Z. Chen & Y. Duan, Complete random forest based class noise filtering learning for improving the generalizability of classifiers. IEEE Transactions on Knowledge and Data Engineering, 2018. 31(11):p.2063-2078.
• 4. ASalman, H. A., A. Kalakech & A. Steiti, Random forest algorithm overview. Babylonian Journal of Machine Learning, 2024. 2024:p. 69-79.
• 5. Montesinos López, O. A., A. Montesinos López & J. Crossa, Support vector machines and support vector regression. In Multivariate statistical machine learning methods for genomic prediction, 2022, Springer International Publishing. p. 337-378.
• 6. Bentéjac, C., A. Csörgő & G. Martínez-Muñoz, A comparative analysis of gradient boosting algorithms. Artificial Intelligence Review, 2021. 54(3): p.1937-1967.
• 7. Talebpour, N., & M. Ilbeigi, Modeling and forecasting uncertainties in power exchange between distributed energy systems and urban grid networks. Energy Reports, 2025. 14: p. 3277-3285.
• 8. Loizidis, S., A. Livera, A. Kyprianou & G. E. Georghiou, Classifying Tomorrow’s Currents: A Probabilistic Neural Network Approach to Forecasting Electricity Prices. In 2024 IEEE 3rd Internatonal Conference on Energy Transition in the Mediterranean Area (SyNERGY MED). 2024, October, p. 1-5.
• 9. Lin, Q., W. Chen, X. Zhao, S. Zhou, X. Gong & B. Zhao, Research on a price prediction model for a multi-layer spot electricity market based on an intelligent learning algorithm. Frontiers in Energy Research, 2024. 12: p. 1308806.
• 10. Masood, Z., R. Gantassi & Y. Choi, Enhancing short-term electric load forecasting for households using quantile LSTM and clustering-based probabilistic approach. IEEE Access, 2024. 12: p.77257-77268.
• 11. Bilgili, M, N. Arslan, A. Şekertekin & A. Yaşar, Application of long short-term memory (LSTM) neural network based on deeplearning for electricity energy consumption forecasting. Turkish Journal of Electrical Engineering and Computer Sciences, 2022. 30 (1): p.140-157.
• 12. O’Connor, C., M. Bahloul, S. Prestwich & A. Visentin, A Review of Electricity Price Forecasting Models in the Day-Ahead, Intra-Day, and Balancing Markets. Energies, 2025. 18(12): p. 3097.
• 13. Laitsos, V., G. Vontzos, P. Paraschoudis, E. Tsampasis, D. Bargiotas & L. H. Tsoukalas, The state of the art electricity load and price forecasting for the modern wholesale electricity market. Energies, 2024. 17(22): p. 5797.
• 14. Li, C., Z. Liu, G. Zhang, Y. Sun, S. Qiu, S. Song & D. Wang, Day-Ahead Electricity Price Forecasting for Sustainable Electricity Markets: A Multi-Objective Optimization Approach Combining Improved NSGA-II and RBF Neural Networks, Sustainability, 2025. 17(10): p. 4551.
• 15. Boru İpek, A. Prediction of market-clearing price using neural networks based methods and boosting algorithms. International Advanced Researches and Engineering Journal, 2021. 5(2) :p. 240-246.
• 16. Watermeyer, M., T. Möbius, O. Grothe and F. Müsgens, A hybrid model for day-ahead electricity price forecasting: Combining fundamental and stochastic modelling, arXiv preprint, arXiv:2304.09336, 2023. https://doi.org/10.48550/arXiv.2304.09336
• 17. Yang, Y., W. Li, T. A. Gulliver & S. Li, Bayesian deep learning-based probabilistic load forecasting in smart grids. IEEE Transactions on Industrial Informatics, 2019. 16(7): p. 4703-4713.
• 18. Deng, X., H. Shao, C. Hu, D. Jiang & Y. Jiang, Wind power forecasting methods based on deep learning: A survey. Computer Modeling in Engineering & Sciences, 2020. 122(1): p. 273-302.
• 19. Devaraj, J., R. Madurai Elavarasan, G. M. Shafiullah, T. Jamal & I. Khan, A holistic review on energy forecasting using big data and deep learning models. International Journal of Energy Research, 2021. 45(9): p. 13489-13530.
• 20. EPİAŞ Transparency Platform, Market Data Portal. [cited 2025 15 November]; Available from: https://seffaflik.epias.com.tr.
• 21. EPİAŞ Transparency Platform, Real-Time Generation Data, [cited 2025 15 November]; Available from: https://seffaflik.epias.com.tr.
• 22. Lin, J.; J. Ma, J. Zhu, Y. Cui, Short-term load forecasting based on LSTM networks considering attention mechanism. Int. J. Electr. Power Energy Syst. 2022. 137,107818.
• 23. Chai T., R.R. Draxler, Root mean square error (RMSE) or mean absolute error (MAE)? Arguments against avoiding RMSE in the literature. Geosci Model Dev, 2014, 7(3): p. 1247–1250.
• 24. Laio, F., S. Tamea, Verification tools for probabilistic forecasts of continuous hydrological variables Hydrol. Earth Syst. Sci. Discuss., 2007, 11 (4): p. 1267-1277.
• 25. Nowotarski, J., R. Weron, Recent advances in electricity price forecasting: A review of probabilistic forecasting Renew. Sustain. Energy Rev., 2018. 81: p. 1548-1568.