Electricity Price Forecasting using Sale and Purchase Curves: The X-Model Evidence for Spain

Abstract

In this Master Thesis, we implement in the Spanish electricity market the X-Model proposed by Ziel and Steinert (2016), combining the insights of market structure models with econometric techniques. Instead of directly modeling the electricity price using past electricity prices, we model and utilize its true source: the sale and purchase curves of the electricity market. We incorporate not only several known features, such as seasonal behavior or renewable energy participation, but also other facts of the bidding structure, such as price clustering. The model is able to capture the non-linear behavior of the electricity price, which is especially useful for predicting huge price spikes. We simulate the months of April 2019 and 2020 and we observe that the model performs better for demand than for supply. We also observe that the supply curve forecast is sensitive to price spikes and that the fit improves as the renewable participation increases. From the policy perspective, this Master Thesis can help to identify the effects of supply-induced changes (such as fossil fuel price changes, the phase-out of coal or nuclear plants or the integration of more renewables) and demand-induced changes (such as the effect of the Covid-19 pandemic). All these issues are in the current energy debate on energy economics.