Abstract
Efficient modeling and forecasting of electricity prices are essential in today’s competitive electricity markets. However, price forecasting is not easy due to the specific features of the electricity price series. This study examines the performance of an ensemble-based technique for forecasting short-term electricity spot prices in the Italian electricity market (IPEX). To this end, the price time series is divided into deterministic and stochastic components. The deterministic component that includes long-term trends, annual and weekly seasonality, and bank holidays, is estimated using semi-parametric techniques. On the other hand, the stochastic component considers the short-term dynamics of the price series and is estimated by time series and various machine learning algorithms. Based on three standard accuracy measures, the results indicate that the ensemble-based model outperforms the others, while the random forest and ARMA are highly competitive.
Highlights
Before the liberalization of the electricity sector, the electrical industry was fully controlled by utility companies, generally state-owned
MODELING AND FORECASTING STOCHASTIC COMPONENT Following the estimation of the deterministic components“using a semi-parametric approach, the stochastic component δd,j is modeled using a combination of autoregressive moving average (ARMA) and different machine learning models, such as neural network autoregressive (NAR), random forest (RF), support vector regression (SVR), gradient boosting machine (GBM), and an ensemble model based on these methods
The forecasting problem of electricity prices in the recently liberalized market is analyzed in depth in this work.“In competitive electricity markets, efficient modeling and forecasting of electricity prices are essential as forecasts are necessary for risk management, trading, and future planning because prices and demand are determined one day before physical delivery
Summary
Before the liberalization of the electricity sector, the electrical industry was fully controlled by utility companies, generally state-owned. To forecast next-day electricity prices, [17] proposed two modeling strategies, i.e., dynamic regression and transfer function They used Spanish and Californian market data to evaluate the performance of their proposed methodology. None of these works apply the ensemble approach in the context of component estimation techniques.“in general, they do not provide any inferential analysis to test differences in the prediction accuracy of the considered models. The forecasting performance is evaluated on a whole year, and the significance analysis of the differences in prediction accuracy is investigated.”In addition, the proposed model can capture the specific features of the electricity price series, leading to higher forecasting accuracy gain.
Sign-in/Register to view highlights & summary 
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
