Online Program Home
My Program

Abstract Details

Activity Number: 426 - Contributed Poster Presentations:Business and Economic Statistics Section
Type: Contributed
Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
Sponsor: Business and Economic Statistics Section
Abstract #333141
Title: Modelling the autoregressive behavior of the European electricity day-ahead markets
Author(s): Rune Hjorth Nielsen* and Ines Wilms and David Matteson
Companies: Aalborg University and KU Leuven and Cornell University

A common approach to modelling the European day-ahead market of the electricity prices is to estimate a vector autoregressive (VAR) model. The infrastructure of these markets entails that 24 variables must be included in the model, which in combination with the usual lag order of at least 7, entails that shrinkage methods are usually applied. These shrinkage methods do not necessarily fit the VAR model the best as they do not consider the dynamic structure of the model. In this poster we present methods that exploit this dynamic model structure by applying structured sparsity penalties on the VAR parameters. This method is compared to different expert models of electricity prices for several European countries, showing the impact of different market infrastructures on the estimated results.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program