The electricity market has been widely introduced in many countries all over the world and the study on electricity price forecast technology has drawn a lot of attention. In this paper, with different parameter Ci and εi assigned to each training data, the flexible Ci Support Vector Regression (SVR) model is developed in terms of the particularity of the price forecast in electricity market. For Day Ahead Market (DAM) price forecast, the load, time of use index and index of day type are taken as the major factors to characterize the market price, therefore, they are selected as the inputs for the flexible SVR forecast model. For the long-term price forecast, we take the reserve margin Rm, HHI and the fuel price index as the inputs, since they are the major factors that drive the market price variation in long run. For short-term price forecast, besides the detailed analysis with the young Italian electricity market, the new model is tested on the experimental stage of the Spanish market, the New York market and the New England market. The long-term forecast with the SVR model presented is justified by the forecast with the data from the Long Run Market Simulator (LREMS).

Price forecast in the competitive electricity market by support vector machine / Gao, C.; Bompard, Ettore Francesco; Napoli, Roberto; Cheng, H.. - In: PHYSICA. A. - ISSN 0378-4371. - 382 (1):(2007), pp. 98-113. (Intervento presentato al convegno 5th International Conference Applications of Physics in Financial Analysis (APFA 5) tenutosi a Torino (ITA) nel June 29th – July 1st , 2006.).

Price forecast in the competitive electricity market by support vector machine

BOMPARD, Ettore Francesco;NAPOLI, Roberto;
2007

Abstract

The electricity market has been widely introduced in many countries all over the world and the study on electricity price forecast technology has drawn a lot of attention. In this paper, with different parameter Ci and εi assigned to each training data, the flexible Ci Support Vector Regression (SVR) model is developed in terms of the particularity of the price forecast in electricity market. For Day Ahead Market (DAM) price forecast, the load, time of use index and index of day type are taken as the major factors to characterize the market price, therefore, they are selected as the inputs for the flexible SVR forecast model. For the long-term price forecast, we take the reserve margin Rm, HHI and the fuel price index as the inputs, since they are the major factors that drive the market price variation in long run. For short-term price forecast, besides the detailed analysis with the young Italian electricity market, the new model is tested on the experimental stage of the Spanish market, the New York market and the New England market. The long-term forecast with the SVR model presented is justified by the forecast with the data from the Long Run Market Simulator (LREMS).
2007
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/1522883
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo