Decomposing electricity prices with jumps

Abstract

We propose a model that decomposes electricity prices into two independent stochastic processes: one that represents the “normal” pattern of electricity prices and another that captures temporary shocks or “jumps”, with non-lasting effects in the market. Each contains specific mean reverting parameters to estimate. In order to identify such components we specify a state-space model with regime switching and apply the Kim’s (1994) filtering algorithm to estimate the model for the mean adjusted series of New South Wales’ electricity prices. Finally, bootstrap simulations were performed to estimate the expected contribution of each of the components in the overall electricity prices.