A longitudinal model for MIBEL energy prices

We propose to contribute to the problematic of Electricity Price Forecasting with a longitudinal statistical approach. We focus our interest on forecasting intra-day prices using hourly data (disaggregated data) in a multivariate approach rather than in the usually used univariate approach, by adjusting a mixed-effects longitudinal model to the Iberian Electricity Market hourly prices from January 1th 2015 to June 26th 2016, in a total of 13 032 observations. Results indicate that a longitudinal approach considering a mixed-effects model, with month and weekday as fixed effects, hour group as random effect and an AutoRegressive component of order 7 describing the within hour dependence, yield a model that explains the intra-day and intra-hour dynamics for the electricity hourly prices.