Please use this identifier to cite or link to this item:
Title: Electricity futures price models : Calibration and forecasting
Authors: Date, P
Islyaev, S
Keywords: Electricity derivatives;Jump diffusion models
Issue Date: 2015
Publisher: Elsevier
Citation: European Journal of Operational Research, 247(1): 144–154, (16 November 2015)
Abstract: A new one factor model with a random volatility parameter is presented in this paper for pricing of electricity futures contracts. It is shown that the model is more tractable than multi-factor jump diffusion models and yields an approximate closed-form pricing formula for the electricity futures prices. On real market data, it is shown that the performance of the new model compares favorably with two existing models in the literature, viz. a two factor jump diffusion model and its jump free version, i.e., a two factor linear Gaussian model, in terms of ability to predict one day ahead futures prices. Further, a multi-stage procedure is suggested and implemented for calibration of the two factor jump diffusion model, which alleviates the difficulty in calibration due to a large number of parameters and pricing formulae which involve numerical evaluation of integrals. We demonstrate the utility of our new model, as well as the utility of the calibration procedure for the existing two factor jump diffusion model, by model calibration and price forecasting experiments on three different futures price data sets from Nord pool electricity data. For the jump diffusion model, we also investigate empirically whether it performs better in terms of futures price prediction than a corresponding, jump-free linear Gaussian model. Finally, we investigate whether an explicit calibration of jump risk premium in the jump diffusion model adds value to the quality of futures price prediction. Our experiments do not yield any evidence that modelling jumps leads to a better price prediction in electricity markets.
ISSN: 0377-2217
Appears in Collections:Dept of Mathematics Research Papers

Files in This Item:
File Description SizeFormat 
Fulltext.pdf293.04 kBAdobe PDFView/Open

Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.