Serviços Personalizados
Journal
Artigo
Indicadores
- Citado por SciELO
- Acessos
Links relacionados
- Citado por Google
- Similares em SciELO
- Similares em Google
Compartilhar
Cuadernos de Administración
versão impressa ISSN 0120-3592
Resumo
VELASQUEZ H, Juan David; FRANCO C, Carlos Jaime e OLAYA M, Yris. Forecasting average monthly prices for Colombian wholesale electricity market contracts. Cuad. Adm. [online]. 2010, vol.23, n.40, pp.321-337. ISSN 0120-3592.
Forecasting electricity prices in liberalized, deregulated markets has always been considered a difficult task, due to the number of factors that govern prices and to their complexity. This article predicts the average monthly prices for Colombian electricity market contracts by using a novel neural network known as the support vector machine. Forecasts obtained using a multilayer perceptron are compared to forecasts obtained using an ARIMA model. The results show that the support vector machine better captures the intrinsic dynamics of the time series and is able to make more precise forecasts considering a 12-month horizon.
Palavras-chave : Comparative studies; non-linear series; prediction; electricity prices; neural networks.