ÇOKLU REGRESYON, ARIMA VE YAPAY SİNİR AĞI YÖNTEMLERİ İLE TÜRKİYE ELEKTRİK PİYASASINDA FİYAT TAHMİN VE ANALİZİ
Abstract
Türkiye’de EPİAŞ aracılığıyla işletilen ve operasyon süreçlerinden birisi olarak elektrik piyasalarında arza çıkan her birim elektrik için arz ve talep dengesine göre piyasa takas fiyatı oluşmaktadır. Piyasa takas fiyatı saatlik olarak oluşmakta katılımcılara ve halka açık olarak şeffaf platformlarda bildirilmektedir. Bu çalışmada R istatistik paket programı kullanılarak çoklu regresyon yöntemi ve yapay sinir ağı modelleri; Eviews paket programı ile ise ARIMA yöntemi kullanılarak geçmiş piyasa fiyatının analizi ve tahmini gerçekleştirilmiştir. Çoklu regresyon ve yapay sinir ağları yöntemleri ile piyasa fiyatını doğrudan etkileyen etmenler bağımsız değişkenler olarak ele alınmıştır. ARIMA yönteminde ise verinin geçmiş değerleri referans olarak alınarak analizler gerçekleştirilmiştir. Çalışmada kullanılan analiz yöntemleri sonucunda en iyi performansın sırasıyla yapay sinir ağları, çoklu regresyon yöntemi ve ARIMA yönteminden elde edildiği belirlenmiştir.
Keywords
Thanks
Değerlendirmeleriniz için teşekkürler.
References
- Akaike, H. (1974). “A New Look at the Statistical Model Identification”, IEEE Transactions on Automatic Control, 19(6), 716-723.
- Carpio, K. J. E., Go, A. M. L., ve Roncal, C. K. M. (2012). “Forecasting Day-Ahead Electricity Prices of Singapore through ARIMA and Wavelet-ARIMA”, DLSU Business & Economic Review, 22(1), 97-118.
- Conejo, A .J., Contreas J., Espinola R., ve Plazas, M. A (2005). “Forecasting Electricity Prices for A Day-Ahead Pool-Based Electric Energy Market”, International Journal of Forecasting, 435-462.
- Çetintaş H. ve Bicil İ.M. (2015). “Elektrik Piyasalarında Yeniden Yapılanma ve Türkiye Elektrik Piyasasında Yapısal Dönüşüm”, Optimum Ekonomi ve Yönetim Bilimleri Dergisi, 2(2), 1-15.
- Dickey, D. A., ve Fuller, W. A., (1979). “Distribution of the Estimators for Autoregressive Time Series with A Unit Root”, Journal of the American Statistical Association, 74(366), 427.
- Ferreira, A. P., Gonçalves, R. J., ve Odete, F, P. (2019). “A Linear Regression Pattern for Electricity Price Forecasting in The Iberian Electricity Market”, Revista Facultad De Ingeniería Universidad De Antioquia, 93, 117-127.
- Gershenson, C. (2003). “Artificial Neural Networks for Beginners”, University of Sussex. 1-8.
- Kalaycı, Ş. (2014). “SPSS Uygulamalı Çok Değişkenli İstatistik Teknikleri (6. baskı)”, Ankara: Asil Yayın Dağıtım.
- Kalfa, V. R., Arslan, B.ve Ertuğrul, İ. (2021). “Determining the Factors Affecting the Market Clearing Price by Using Multiple Linear Regression Method” . Alphanumeric Journal , 9(1) , 35-48
- Kaya vd. (2012). “Karışımlı İkili Lojistik Regresyon Modeline İlişkin Bir Uygulama”, DEÜ Mühendislik Fakültesi Mühendislik Bilimleri Dergisi, 14(1), 39-47.
- Khare, M. ve Nagendra, S. M. S. (2006). “Artificial Neural Networks in Vehicular Pollution Modelling”, Studies in Computational Intelligence, Poland.
- Kölmek, M. A. ve Navruz, İ. (2012). “Forecasting The Day-Ahead Price in Electricity Balancing and Settlement Market of Turkey by Using Artificial Neural Networks”, Turkey Journal of Electrical Engineering & Computer Sciences, 23, 841–852.
- Onaiwu E. (2009). “How Does Bilateral Trading Differ from Electricity Pooling”, University of Dundee.
- Ozozen, A., Kayakutlu G., Ketterer M., ve Kayalica, O. (2016). “A Combined Seasonal ARIMA and ANN Model for İmproved Results in Electricity Spot Price Forecasting: Case Study in Turkey”, 2016. Proceedings of PICMET '16: Technology Management for Social Innovation, 2681-2690.
- Öztemel, E. (2006). “Yapay Sinir Ağları”, İstanbul: Papatya Yayıncılık
- Peter, D. ve Silvia P. (2012). “ARIMA Vs. ARIMAX – Which Approach is Better to Analyze and Forecast Macroeconomic Time Series”, Proceedings of 30th International Conference Mathematical Methods in Economics, 136-140.
- Torró, H. (2007), “Forecasting Weekly Electricity Prices at Nord Pool”. Working Papers 2007.88, Fondazione Eni Enrico Mattei.
- Yıldıran, A. ve Kandemir, S. Y. (2018). “Yağış Miktarının Yapay Sinir Ağları ile Tahmini”, BŞEÜ Fen Bilimleri Dergisi, 5(2), 97-104.
- Werbos, P. J. (1974). “Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences”, Phd. Thesis, Harvard University.
FORECASTING AND ANALYSIS IN TURKISH ELECTRICITY MARKET WITH MULTIPLE REGRESSION, ARIMA AND ARTIFICIAL NEURAL NETWORK METHODS
Abstract
For each unit of electricity supplied in electricity markets, one of the operation processes and operated through EPİAŞ in Turkey, a market-clearing price is formed according to the supply and demand balance. The market-clearing price is formed hourly and reported to the participants on transparent platforms. The aim of this study is to analyse the market-clearing price in Turkish Electricity Markets and to compare the performances of statistical prediction analysis methods frequently used in the literature. In this study, multiple regression method and artificial neural networks models were applied by using R statistical package program; with the Eviews package program the analysis and estimation of the past market price was performed using the ARIMA method. Analyses were applied to 2928 data of market clearing prices realized in March, April, May and June 2020. In the multiple regression and artificial neural network models, factors, thought to directly affect the market price, are considered as independent variables. In the ARIMA method, analyses were carried out by taking the past values of the data as a reference. The results of the analysis methods used in the study show that the best performance was obtained through artificial neural networks, multiple regression method and ARIMA method, respectively.
References
- Akaike, H. (1974). “A New Look at the Statistical Model Identification”, IEEE Transactions on Automatic Control, 19(6), 716-723.
- Carpio, K. J. E., Go, A. M. L., ve Roncal, C. K. M. (2012). “Forecasting Day-Ahead Electricity Prices of Singapore through ARIMA and Wavelet-ARIMA”, DLSU Business & Economic Review, 22(1), 97-118.
- Conejo, A .J., Contreas J., Espinola R., ve Plazas, M. A (2005). “Forecasting Electricity Prices for A Day-Ahead Pool-Based Electric Energy Market”, International Journal of Forecasting, 435-462.
- Çetintaş H. ve Bicil İ.M. (2015). “Elektrik Piyasalarında Yeniden Yapılanma ve Türkiye Elektrik Piyasasında Yapısal Dönüşüm”, Optimum Ekonomi ve Yönetim Bilimleri Dergisi, 2(2), 1-15.
- Dickey, D. A., ve Fuller, W. A., (1979). “Distribution of the Estimators for Autoregressive Time Series with A Unit Root”, Journal of the American Statistical Association, 74(366), 427.
- Ferreira, A. P., Gonçalves, R. J., ve Odete, F, P. (2019). “A Linear Regression Pattern for Electricity Price Forecasting in The Iberian Electricity Market”, Revista Facultad De Ingeniería Universidad De Antioquia, 93, 117-127.
- Gershenson, C. (2003). “Artificial Neural Networks for Beginners”, University of Sussex. 1-8.
- Kalaycı, Ş. (2014). “SPSS Uygulamalı Çok Değişkenli İstatistik Teknikleri (6. baskı)”, Ankara: Asil Yayın Dağıtım.
- Kalfa, V. R., Arslan, B.ve Ertuğrul, İ. (2021). “Determining the Factors Affecting the Market Clearing Price by Using Multiple Linear Regression Method” . Alphanumeric Journal , 9(1) , 35-48
- Kaya vd. (2012). “Karışımlı İkili Lojistik Regresyon Modeline İlişkin Bir Uygulama”, DEÜ Mühendislik Fakültesi Mühendislik Bilimleri Dergisi, 14(1), 39-47.
- Khare, M. ve Nagendra, S. M. S. (2006). “Artificial Neural Networks in Vehicular Pollution Modelling”, Studies in Computational Intelligence, Poland.
- Kölmek, M. A. ve Navruz, İ. (2012). “Forecasting The Day-Ahead Price in Electricity Balancing and Settlement Market of Turkey by Using Artificial Neural Networks”, Turkey Journal of Electrical Engineering & Computer Sciences, 23, 841–852.
- Onaiwu E. (2009). “How Does Bilateral Trading Differ from Electricity Pooling”, University of Dundee.
- Ozozen, A., Kayakutlu G., Ketterer M., ve Kayalica, O. (2016). “A Combined Seasonal ARIMA and ANN Model for İmproved Results in Electricity Spot Price Forecasting: Case Study in Turkey”, 2016. Proceedings of PICMET '16: Technology Management for Social Innovation, 2681-2690.
- Öztemel, E. (2006). “Yapay Sinir Ağları”, İstanbul: Papatya Yayıncılık
- Peter, D. ve Silvia P. (2012). “ARIMA Vs. ARIMAX – Which Approach is Better to Analyze and Forecast Macroeconomic Time Series”, Proceedings of 30th International Conference Mathematical Methods in Economics, 136-140.
- Torró, H. (2007), “Forecasting Weekly Electricity Prices at Nord Pool”. Working Papers 2007.88, Fondazione Eni Enrico Mattei.
- Yıldıran, A. ve Kandemir, S. Y. (2018). “Yağış Miktarının Yapay Sinir Ağları ile Tahmini”, BŞEÜ Fen Bilimleri Dergisi, 5(2), 97-104.
- Werbos, P. J. (1974). “Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences”, Phd. Thesis, Harvard University.
Details
| Primary Language | Turkish |
|---|---|
| Journal Section | Articles |
| Authors | |
| Publication Date | March 22, 2022 |
| Published in Issue | Year 2022 Volume: 20 Issue: 1 |