MCMC estimation of a multi-factor jump diffusion model for power prices

Rikard Green, Marcus Nossman

Abstract

ABSTRACT

In this paper we generalize the electricity spot price model of Lucia and Schwartz by a two-factor model with jumps and stochastic volatility.We estimate the model on daily spot price data from the Nordic market using an approach that combines traditional statistical methods with a Markov chain Monte Carlo algorithm. Results show that the model captures most of the trajectorial and the statistical characteristics of the electricity spot price. Further, we find that inclusion of stochastic volatility is crucial to separate spikes from the normal price process. Moreover, we estimate that the correlation between the spot price and its stochastic volatility is negative.