Title :
Forecasting price spikes in European day-ahead electricity markets using decision trees
Author :
Fragkioudaki, Anna ; Marinakis, Adamantios ; Cherkaoui, Rachid
Author_Institution :
Univ. of Seville, Seville, Spain
Abstract :
Although the techniques for normal price prediction are well advanced, price spike prediction remains a challenge. The aim of this paper is to present a method to predict both normal day-ahead electricity prices as well as day-ahead price spikes using classification and regression trees. The models are build using historical data of European electricity market and transmission system variables. The proposed method is tested on several bidding zones of Europe.
Keywords :
decision trees; economic forecasting; pattern classification; power markets; regression analysis; tendering; European bidding zone; European day-ahead electricity market; European electricity market; European transmission system; classification tree; decision tree; normal price prediction technique; price spike forecasting; price spike prediction; regression tree; Electricity supply industry; Europe; Forecasting; Power systems; Predictive models; Time series analysis; Training; Day-ahead electricity markets; decision trees; feature selection; price forecasting; price spikes;
Conference_Titel :
European Energy Market (EEM), 2015 12th International Conference on the
Conference_Location :
Lisbon
DOI :
10.1109/EEM.2015.7216672
