Sparse time-varying parameter VECMs with an application to modeling electricity prices

TL;DR

This paper introduces a novel sparse, time-varying parameter VECM with automatic model specification, applied to European electricity prices, improving forecasting accuracy by addressing cointegration and nonlinearities.

Contribution

It develops a new sparse TVP VECM framework with automatic model selection and applies it to electricity prices, capturing cointegration and nonlinear effects.

Findings

Model achieves competitive forecast accuracy for hourly German electricity prices.

Explicitly modeling cointegration improves understanding of price dynamics.

Sparse solutions reduce overfitting and parameter uncertainty.

Abstract

In this paper we propose a time-varying parameter (TVP) vector error correction model (VECM) with heteroskedastic disturbances. We propose tools to carry out dynamic model specification in an automatic fashion. This involves using global-local priors, and postprocessing the parameters to achieve truly sparse solutions. Depending on the respective set of coefficients, we achieve this via minimizing auxiliary loss functions. Our two-step approach limits overfitting and reduces parameter estimation uncertainty. We apply this framework to modeling European electricity prices. When considering daily electricity prices for different markets jointly, our model highlights the importance of explicitly addressing cointegration and nonlinearities. In a forecast exercise focusing on hourly prices for Germany, our approach yields competitive metrics of predictive accuracy.