Forecasting Electricity Prices with Expert, Linear and Non-Linear Models

This paper provides an iterative model selection for forecasting day–ahead hourly electricity prices, while accounting for fundamental drivers. Forecasts of demand, in-feed from renewable energy sources (RES), fossil fuel prices, and physical flows are all included in linear and nonlinear specifications, ranging in the class of ARFIMA–GARCH models hence including parsimonious autoregressive specifications (known as expert-type models). Results support the adoption of a simple structure that is able to adapt to market conditions. Indeed, we include forecasted demand, wind and solar power, actual generation from hydro, biomass and waste, weighted imports and traditional fossil fuels. The inclusion of these exogenous regressors, in both the conditional mean and variance equations, outperforms in point and, especially, in density forecasting. Considering the northern Italian prices and using the Model Confidence Set, predictions indicate a strong predictive power of regressors, in particular in an expert model augmented for GARCH-type time-varying volatility. Finally, we find that using professional and more timely predictions of consumption and RES improves the forecast accuracy of electricity prices more than predictions freely available to researchers.