Comparing the Forecasting Performances of Linear Models for Electricity Prices with High RES Penetration

TL;DR

This study evaluates various statistical models for forecasting hourly electricity prices in European markets with high renewable energy, finding Bayesian VAR models with exogenous variables to be most effective.

Contribution

It provides a comprehensive comparison of univariate and multivariate, frequentist and Bayesian models for electricity price forecasting in renewable-rich markets.

Findings

Bayesian VAR with exogenous variables outperforms other models.

Multivariate Bayesian models yield more accurate forecasts.

Model performance varies across different European markets.

Abstract

This paper compares alternative univariate versus multivariate models, frequentist versus Bayesian autoregressive and vector autoregressive specifications, for hourly day-ahead electricity prices, both with and without renewable energy sources. The accuracy of point and density forecasts are inspected in four main European markets (Germany, Denmark, Italy and Spain) characterized by different levels of renewable energy power generation. Our results show that the Bayesian VAR specifications with exogenous variables dominate other multivariate and univariate specifications, in terms of both point and density forecasting.