Authors: Konstantinia Daskalou, Christina Diakaki

Title: Studying the day ahead electricity price forecasting problem in coupled markets: The case of Italy

Abstract

Day ahead electricity price forecasting is an extensively studied problem, and several statistical, intelligence-based and other techniques have been proposed in literature to address it. However, the liberalization of the electricity market taking place during the last decades, and the market coupling pursued within the European Union reshape the problem and create the need to confirm the effectiveness and/or revise existing methods and solution techniques, and/or invent new approaches. Given that complete integration has not achieved yet, both relevant data and studies of forecasting considering integration are still rather sparse. It has thus been the aim of this study to contribute in filling this gap by focusing on and investigating the market integration effects in day ahead electricity price forecasting. To this end, an Artificial Neural Network was developed, and used under several, with respect to inputs, forecasting scenarios considering the Italian electricity market. The performed investigations confirmed like other past studies that the consideration of market coupling and/or interconnection may indeed improve the day ahead electricity price forecasting accuracy. They also, however, suggested that the geographical proximity of considered markets may affect the forecasting results; the closer the markets, the stronger the effects. Thus market coupling effects need further investigation.