Comparison of indicator saturation and Markov regime-switching models for Brazilian electricity prices

Abstract

In this study, we compared the results obtained from the application of two different approaches to model Brazilian electricity prices. One of them is the Indicator Saturation model, which consider simultaneously impulse, step, trend events and autocorrelation. The other one is the Markov regime-switching model. We considered the marginal operating cost as the Brazilian electricity spot price, and the study was applied in the four different Brazilian submarkets. We concluded that the Indicator Saturation approach outperforms the regime-switching model in terms of outlier detection. We discussed the relevance of this study as a response to the increase in electricity price volatility caused by the energy transition, as well as to enrich the debate related to the possible change in the methodology for calculating the Brazilian electricity spot price within the framework of the electricity sector reform underway in Brazil.