2008, Vol.11, No.2, pp.215-224

Power prices dynamics in deregulated markets appears variable and unpredictable with jumps, spikes, and high, non constant, volatility. Empirical distributions of log-returns are characterized by large values of the standard deviation as well as non-zero skewness and very high kurtosis. In this paper we discuss a reduced-form methodology to describe the dynamics of electricity prices in order to capture the statistical properties observed in real markets. Particular attention will be devoted to regime-switching models which seem good candidates to incorporate the main features of power prices as the seasonality component, the occurrence of stable and turbulent periods, as well as jumps and spikes. Regime-switching models offer, indeed, the possibility to introduce various mean-reversion rates and volatilities depending on the state of the system thus enhancing the flexibility of the reduced-form approach. An empirical analysis performed on market data is provided to test the adaptability of the discussed models in replicating the first four moment of the empirical log-returns distributions.
Key words: nonlinear dynamics, regime-switching model, market

Full text:  Acrobat PDF  (358KB)  

Copyright © Nonlinear Phenomena in Complex Systems. Last updated: July 10, 2008