Seasonality in deep learning forecasts of electricity imbalance prices

Sinan Deng; John Inekwe; Vladimir Smirnov; Andrew Wait; Chao Wang

doi:10.1016/j.eneco.2024.107770

Seasonality in deep learning forecasts of electricity imbalance prices

Sinan Deng, John Inekwe, Vladimir Smirnov, Andrew Wait^*, Chao Wang

^*Corresponding author for this work

Department of Actuarial Studies and Business Analytics

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

10 Downloads (Pure)

Abstract

In this paper, we propose a seasonal attention mechanism, the effectiveness of which is evaluated via the Bidirectional Long Short-Term Memory (BiLSTM) model. We compare its performance with alternative deep learning and machine learning models in forecasting the balancing settlement prices in the electricity market of Great Britain. Critically, the Seasonal Attention-Based BiLSTM framework provides a superior forecast of extreme prices with an out-of-sample gain in the predictability of 11%–15% compared with models in the literature. Our forecasting techniques could aid both market participants, to better manage their risk and assign their assets, and policy makers, to operate the system at lower cost.

Original language	English
Article number	107770
Pages (from-to)	1-10
Number of pages	10
Journal	Energy Economics
Volume	137
DOIs	https://doi.org/10.1016/j.eneco.2024.107770
Publication status	Published - Sept 2024

Bibliographical note

© 2024 The Authors. Published by Elsevier B.V. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

Balance settlement prices
Deep learning
Electricity
Forecasting
Machine learning

Access to Document

10.1016/j.eneco.2024.107770Licence: CC BY

Publisher version (open access)Final published version, 2.22 MBLicence: CC BY

Cite this

@article{881e2a22444f4b39974703a58b54d802,

title = "Seasonality in deep learning forecasts of electricity imbalance prices",

abstract = "In this paper, we propose a seasonal attention mechanism, the effectiveness of which is evaluated via the Bidirectional Long Short-Term Memory (BiLSTM) model. We compare its performance with alternative deep learning and machine learning models in forecasting the balancing settlement prices in the electricity market of Great Britain. Critically, the Seasonal Attention-Based BiLSTM framework provides a superior forecast of extreme prices with an out-of-sample gain in the predictability of 11%–15% compared with models in the literature. Our forecasting techniques could aid both market participants, to better manage their risk and assign their assets, and policy makers, to operate the system at lower cost.",

keywords = "Balance settlement prices, Deep learning, Electricity, Forecasting, Machine learning",

author = "Sinan Deng and John Inekwe and Vladimir Smirnov and Andrew Wait and Chao Wang",

note = "{\textcopyright} 2024 The Authors. Published by Elsevier B.V. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.",

year = "2024",

month = sep,

doi = "10.1016/j.eneco.2024.107770",

language = "English",

volume = "137",

pages = "1--10",

journal = "Energy Economics",

issn = "0140-9883",

publisher = "Elsevier",

}

TY - JOUR

T1 - Seasonality in deep learning forecasts of electricity imbalance prices

AU - Deng, Sinan

AU - Inekwe, John

AU - Smirnov, Vladimir

AU - Wait, Andrew

AU - Wang, Chao

N1 - © 2024 The Authors. Published by Elsevier B.V. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2024/9

Y1 - 2024/9

N2 - In this paper, we propose a seasonal attention mechanism, the effectiveness of which is evaluated via the Bidirectional Long Short-Term Memory (BiLSTM) model. We compare its performance with alternative deep learning and machine learning models in forecasting the balancing settlement prices in the electricity market of Great Britain. Critically, the Seasonal Attention-Based BiLSTM framework provides a superior forecast of extreme prices with an out-of-sample gain in the predictability of 11%–15% compared with models in the literature. Our forecasting techniques could aid both market participants, to better manage their risk and assign their assets, and policy makers, to operate the system at lower cost.

AB - In this paper, we propose a seasonal attention mechanism, the effectiveness of which is evaluated via the Bidirectional Long Short-Term Memory (BiLSTM) model. We compare its performance with alternative deep learning and machine learning models in forecasting the balancing settlement prices in the electricity market of Great Britain. Critically, the Seasonal Attention-Based BiLSTM framework provides a superior forecast of extreme prices with an out-of-sample gain in the predictability of 11%–15% compared with models in the literature. Our forecasting techniques could aid both market participants, to better manage their risk and assign their assets, and policy makers, to operate the system at lower cost.

KW - Balance settlement prices

KW - Deep learning

KW - Electricity

KW - Forecasting

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85199166503&partnerID=8YFLogxK

U2 - 10.1016/j.eneco.2024.107770

DO - 10.1016/j.eneco.2024.107770

M3 - Article

AN - SCOPUS:85199166503

SN - 0140-9883

VL - 137

SP - 1

EP - 10

JO - Energy Economics

JF - Energy Economics

M1 - 107770

ER -

Seasonality in deep learning forecasts of electricity imbalance prices

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Cite this