In this paper, we present and compare two approaches to model supply and demand curves of a sealed-bid auction market. Both the Linear Hinges model and the Sigmo model are able to extract the relevant information from these bidding curves, without losing significant market information. We discuss their main similarities and important differences using an unified framework, to highlight their main strengths and weakness. A practical comparative study based on real supply functions from the Californian electricity market has been included to derive practical conclusions.
Palabras clave: Bidding curves, supply functions, electricity market, curve fitting, data mining, automatic learning, neural networks.
IEEE Power Tech. Conference, Vol. I, POM6-395, ISBN: 0-7803-7139-9, Oporto, Norte (Portugal). 10 septiembre 2001
Fecha de publicación: septiembre 2001.
A. Mateo, E.F. Sánchez-Úbeda, A. Muñoz, J. García-González, J. Villar, M. Casado, A. Sáiz, E.J. García, R. González, Modeling bidding curves: The linear hinges. Model versus the sigmo model, IEEE Power Tech. Conference, Vol. I, POM6-395, ISBN: 0-7803-7139-9. Oporto, Portugal, 10-13 Septiembre 2001