Regime-switching characterization of electricity prices dynamics

Stochastic models of electricity prices have been used extensively during the last few years to describe prices fluctuations in deregulated power markets. Regime-switching models seem good candidates to capture the main features of electricity prices dynamics as the mean-reversion property as well the presence of jumps and spikes. Since they offer the possibility to introduce various mean-reversion rates, volatility and jumps, depending on the state of the system, such models allow to describe the properties of the stable motion and of the spike dynamics in a very flexible way. In this paper, two-regime and three-regime models are discussed, and a comparison performed on market data, is proposed.