Robust forecasting of electricity prices: Simulations, models and the impact of renewable sources

• Published in: Technological forecasting & social change 2019-04, Vol.141, p.305-318
• Main Authors: Grossi, Luigi, Nan, Fany
• Format: Article
• Language: English
• Subjects: Autoregressive models; Computer simulation; Economic forecasting; Economic models; Electricity; Electricity price; Electricity pricing; Forecasting; Nonlinear time series; Nonlinearity; Parameter estimation; Polynomials; Price forecasting; Prices; Regression analysis; Robust GM-estimator; Robustness (mathematics); Spikes; Threshold models; Weighting; Weighting functions
• ISSN: 0040-1625; 1873-5509
• DOI: 10.1016/j.techfore.2019.01.006

In this paper a robust approach to modeling electricity spot prices is introduced. Differently from what has been recently done in the literature on electricity price forecasting, where the attention has been mainly drawn by the prediction of spikes, the focus of this contribution is on the robust estimation of nonlinear SETARX models (Self-Exciting Threshold Auto Regressive models with eXogenous regressors). In this way, parameter estimates are not, or very lightly, influenced by the presence of extreme observations and the large majority of prices, which are not spikes, could be better forecasted. A Monte Carlo study is carried out in order to select the best weighting function for Generalized M-estimators of SETAR processes. A robust procedure to select and estimate nonlinear processes for electricity prices is introduced, including robust tests for stationarity and nonlinearity and robust information criteria. The application of the procedure to the Italian electricity market reveals the forecasting superiority of the robust GM-estimator based on the polynomial weighting function with respect to the non-robust Least Squares estimator. Finally, the introduction of generation from renewable sources in the robust estimation of SETARX processes contributes to the improvement of the forecasting ability of the model. •Very high prices can dramatically bias electricity price predictions.•Robust estimators improve the overall forecasting performance of the models.•Predicted wind generation and predicted demand of electricity improve the forecasting performance by 15–20%.