Covariance structures for high-dimensional energy forecasting

Browell, J. , Gilbert, C. and Fasiolo, M. (2022) Covariance structures for high-dimensional energy forecasting. Electric Power Systems Research, 211, 108446. (doi: 10.1016/j.epsr.2022.108446)

Abstract

Forecasts of various quantities over multiple time periods and/or spatial expanses are required to operate modern power systems. Furthermore, probabilistic forecasts are necessary to facilitate economic decision-making and risk management. This gives rise to the challenge of producing forecasts which capture the dependency between variables, over time, and between locations. The Gaussian Copula has been widely used for multivariate energy forecasts and is scalable because the entire dependency structure is captured by a covariance matrix; estimating this covariance matrix in high dimensional problems remains a research challenge. Here we focus on parametrising this covariance matrix as a step towards more robust estimation and to enable conditioning on explanatory variables. We present a range of parametric structures and estimation strategies suitable for multivariate energy forecasting.

• Item Type: Articles
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Browell, Dr Jethro

Creator Roles

Browell, J.: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review and editing, Visualization, Project administration, Funding acquisition

• Authors: Browell, J., Gilbert, C., and Fasiolo, M.
• College/School: College of Science and Engineering > School of Mathematics and Statistics > Statistics
• Journal Name: Electric Power Systems Research
• Publisher: Elsevier
• ISSN: 0378-7796
• ISSN (Online): 1873-2046
• Published Online: 13 July 2022
• Copyright Holders: Copyright © 2022 The Authors
• First Published: First published in Electric Power Systems Research 211: 108446
• Publisher Policy: Reproduced under a Creative Commons License
• Data DOI: 10.5281/zenodo.5541782

Project Code	Award No	Project Name	Principal Investigator	Funder's Name	Funder Ref	Lead Dept
315958		System-wide Probabilistic Energy Forecasting	Jethro Browell	Engineering and Physical Sciences Research Council (EPSRC)	EP/R023484/2	M&S - Statistics
315958		System-wide Probabilistic Energy Forecasting	Jethro Browell	Engineering and Physical Sciences Research Council (EPSRC)	EP/R023484/2	M&S - Statistics