SavedAdd your journal

Journals / Journal of Energy Systems / 2026 / Volume: 10 Issue: 3

Modeling and predicting provincial per capita electricity consumption in Türkiye: A panel data and machine learning approach

Yasemin Deniz Okur(GAZİ ÜNİVERSİTESİ)0009-0006-0422-0029
Filiz Kardiyen(GAZİ ÜNİVERSİTESİ, FEN FAKÜLTESİ)0000-0002-8730-2751
Pages
—
Publication Date
2026-09-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results indicate that the industrial share of GDP and cooling degree days (CDD) are positively associated with electricity consumption.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study provides a province-level assessment of per capita electricity consumption across all 81 provinces of Türkiye for 2015–2023.
📊
Key Findings & Data:By contrast, the enterprise rate is negatively associated with consumption, reflecting differences in provincial economic structure rather than energy-intensive production.
💡
Conclusion & Impact:The results indicate that the industrial share of GDP and cooling degree days (CDD) are positively associated with electricity consumption.
🔬Design:Modeling & Simulation
📍Scope:2015–2023 Verileri
⚙️Variables:Electricity demand · Machine learning · Panel data analysis

Abstract

This study provides a province-level assessment of per capita electricity consumption across all 81 provinces of Türkiye for 2015–2023. It employs two-way fixed-effects panel estimation alongside Random Forest modeling to examine regional electricity demand, assess the relative importance of economic, demographic, and climate-related factors, explore potential non-linearities, and evaluate machine-learning evidence in relation to panel-based findings. The results indicate that the industrial share of GDP and cooling degree days (CDD) are positively associated with electricity consumption. By contrast, the enterprise rate is negatively associated with consumption, reflecting differences in provincial economic structure rather than energy-intensive production. Urbanization, renewable energy capacity, and natural gas consumption do not appear to have statistically significant effects within the model. The Random Forest results also support these findings. Prediction error decreases by approximately 22% relative to a baseline model that includes only province and year effects, and variable importance measures highlight the prominent role of the industrial share of GDP. Overall, the findings suggest that provincial electricity demand in Türkiye is more closely related to economic structure and cooling needs than to supply-side factors, with implications for more accurate regional electricity demand forecasting.

🔖

Cite This Article

Okur, Y. D., Kardiyen, F. (2026). Modeling and predicting provincial per capita electricity consumption in Türkiye: A panel data and machine learning approach. Journal of Energy Systems, Volume: 10 Issue: 3. https://doi.org/10.30521/jes.1969238

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (24)

  1. Kavaklıoğlu K. Modeling and prediction of Turkey’s electricity consumption using support vector regression. Applied Energy 2011; 88: 368-375, DOI: 10.1016/j.apenergy.2010.07.021.
  2. Nie RX, Tian ZP, Long RY, Dong W. Forecasting household electricity demand with hybrid machine learning-based methods: Effects of residents’ psychological preferences and calendar variables. Expert Systems with Applications 2022; 206: 117854, DOI: 10.1016/j.eswa.2022.117854.
  3. Günay ME. Forecasting annual gross electricity demand by artificial neural networks using predicted values of socio-economic indicators and climatic conditions: Case of Turkey. Energy Policy 2016; 90: 92-101, DOI: 10.1016/j.enpol.2015.12.019.
  4. Turkish Electricity Transmission Corporation (TEİAŞ). Electricity Generation and Transmission Statistics. Available at: https://www.teias.gov.tr/turkiye-elektrik-uretim-iletim-istatistikleri (accessed 24 January 2026).
  5. Kaytez F, Taplamacioglu MC, Cam E, Hardalac F. Forecasting electricity consumption: A comparison of regression analysis, neural networks and least squares support vector machines. International Journal of Electrical Power & Energy Systems 2015; 67: 431-438, DOI: 10.1016/j.ijepes.2014.12.036.
  6. Sözen A, Arcaklıoğlu E. Prediction of net energy consumption based on economic indicators (GNP and GDP) in Turkey. Energy Policy 2007; 35: 4981-4992, DOI: 10.1016/j.enpol.2007.04.029.
  7. Dilaver Z, Hunt LC. Turkish aggregate electricity demand: An outlook to 2020. Energy 2011; 36: 6686-6696, DOI: 10.1016/j.energy.2011.07.043.
  8. Kurt E, Kasap R, Çelik K. Forecasting of monthly electricity generation from conventional and renewable resources following the coronavirus pandemic in Turkey. Journal of Energy Systems 2022; 6: 420-435, DOI: 10.30521/jes.1106313.
  9. Mohamed Z, Bodger P. Forecasting electricity consumption in New Zealand using economic and demographic variables. Energy 2005; 30: 1833-1843, DOI: 10.1016/j.energy.2004.08.012.
  10. Bianco V, Manca O, Nardini S. Electricity consumption forecasting in Italy using linear regression models. Energy 2009; 34: 1413-1421, DOI: 10.1016/j.energy.2009.06.034.
  11. Bui TH, Lee K. Forecasting annual electricity consumption in Vietnam using radial basis function neural network. Energy 2025; 334: 137762, DOI: 10.1016/j.energy.2025.137762.
  12. Al-Qahtani FH, Crone SF. Multivariate k-nearest neighbour regression for time series data: A novel algorithm for forecasting UK electricity demand. In: Proceedings of the International Joint Conference on Neural Networks (IJCNN); 4-9 August 2013; Dallas, TX, USA: pp. 1-8.
  13. Niu D, Wang Y, Wu DD. Power load forecasting using support vector machine and ant colony optimization. Expert Systems with Applications 2010; 37: 2531-2539, DOI: 10.1016/j.eswa.2009.08.019.
  14. Turkish Statistical Institute (TURKSTAT). Provincial statistics on population, enterprise, education and urbanization. Available at: https://www.tuik.gov.tr (accessed 24 January 2026).
  15. Turkish State Meteorological Service. Heating and cooling degree days data. Available at: https://www.mgm.gov.tr (accessed 24 January 2026).
  16. Energy Market Regulatory Authority (EMRA). Electricity and natural gas consumption and renewable energy statistics. Available at: https://www.epdk.gov.tr (accessed 24 January 2026).
  17. Wooldridge JM. Econometric Analysis of Cross Section and Panel Data. Cambridge, MA, USA: MIT Press, 2002.
  18. Adekeye KS, Igwe KE, Olayiwola OM. On pooled OLS and panel regression models for assessing the contributions of electronic payment systems on commercial banks profitability. Journal of Statistics: Advances in Theory and Applications 2021; 25: 61-81, DOI: 10.18642/jsata_7100122206.
  19. Baltagi BH. Econometric Analysis of Panel Data. Chichester, UK: John Wiley & Sons, 2005.
  20. Zulfikar R. Estimation model and selection method of panel data regression: An overview of common effect, fixed effect, and random effect model. INA-Rxiv 2018. Available at: https://osf.io/9qe2b/ (accessed 27 February 2026).
  21. Abdullah H, Sahudin Z, Bahrudin NZ. Short guides to static panel data regression model estimator. Asian Journal of Accounting and Finance 2022; 4: 1-6, DOI: 10.55057/ajafin.2022.4.4.1.
  22. Raza A, Alharthi M, Shakeel M, Khan MI. Nexus between CO₂ emissions, economic growth, and energy consumption with special reference to BRICS: A panel data approach. Energy Nexus 2025; 20: 100556, DOI: 10.1016/j.nexus.2025.100556.
  23. Hoechle D. Robust standard errors for panel regressions with cross-sectional dependence. The Stata Journal 2007; 7: 281-312.
  24. Landowska A, Kłopotek RA, Filip D, Raczkowski K. GDP-GFCF dynamics across global economies: A comparative study of panel regressions and Random Forest. arXiv preprint arXiv:2504.20993, 2025. Available at: https://arxiv.org/abs/2504.20993.