Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles

Ties van der Heijden; Peter Palensky; Edo Abraham

doi:10.1109/IECON48115.2021.9589097

Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles

Ties van der Heijden, Peter Palensky, Edo Abraham

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

2 Citations (Scopus)

79 Downloads (Pure)

Abstract

In this paper we propose a Quantile Regression Deep Neural Network capable of forecasting multiple quantiles in one model using a combined quantile loss function, and apply it to probabilistically forecast the prices of 8 European Day Ahead Markets. We show that the proposed loss function significantly reduces the quantile crossing problem to (near) 0% in all markets considered, while in some cases simultaneously increasing forecasting performance based on classical point forecast metrics applied to the expected value of the probabilistic forecast. The models are optimized using an automated approach with an elaborate feature- and hyperparameter search space, leading to good model performance in all considered markets.

Original language	English
Title of host publication	IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society
Subtitle of host publication	Proceedings
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	978-1-6654-3554-3
ISBN (Print)	978-1-6654-0256-9
DOIs	https://doi.org/10.1109/IECON48115.2021.9589097
Publication status	Published - 2021
Event	IECON 2021: 47th Annual Conference of the IEEE Industrial Electronics Society - Virtual at Toronto, Canada Duration: 13 Oct 2021 → 16 Oct 2021 Conference number: 47th

Publication series

Name	IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY
ISSN (Print)	1553-572X

Conference

Conference	IECON 2021
Country/Territory	Canada
City	Virtual at Toronto
Period	13/10/21 → 16/10/21

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care

Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Quantile Regression
Electricity Price Forecasting
Deep Neural Network
Day Ahead Market
Crossing Quantile Problem

Access to Document

10.1109/IECON48115.2021.9589097

Probabilistic_DAM_price_forecasting_using_a_combined_Quantile_Regression_Deep_Neural_Network_with_less-crossing_quantilesFinal published version, 1.05 MB

Cite this

van der Heijden, T., Palensky, P., & Abraham, E. (2021). Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles. In IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society: Proceedings Article 9589097 (IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY). IEEE. https://doi.org/10.1109/IECON48115.2021.9589097

van der Heijden, Ties ; Palensky, Peter ; Abraham, Edo. / Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles. IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society: Proceedings. IEEE, 2021. (IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY).

@inproceedings{cbcc9c19a70c42ddb671a0bf53de474e,

title = "Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles",

abstract = "In this paper we propose a Quantile Regression Deep Neural Network capable of forecasting multiple quantiles in one model using a combined quantile loss function, and apply it to probabilistically forecast the prices of 8 European Day Ahead Markets. We show that the proposed loss function significantly reduces the quantile crossing problem to (near) 0% in all markets considered, while in some cases simultaneously increasing forecasting performance based on classical point forecast metrics applied to the expected value of the probabilistic forecast. The models are optimized using an automated approach with an elaborate feature- and hyperparameter search space, leading to good model performance in all considered markets.",

keywords = "Quantile Regression, Electricity Price Forecasting, Deep Neural Network, Day Ahead Market, Crossing Quantile Problem",

author = "{van der Heijden}, Ties and Peter Palensky and Edo Abraham",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; IECON 2021 : 47th Annual Conference of the IEEE Industrial Electronics Society ; Conference date: 13-10-2021 Through 16-10-2021",

year = "2021",

doi = "10.1109/IECON48115.2021.9589097",

language = "English",

isbn = "978-1-6654-0256-9",

series = "IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY",

publisher = "IEEE",

booktitle = "IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society",

address = "United States",

}

van der Heijden, T , Palensky, P & Abraham, E 2021, Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles. in IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society: Proceedings., 9589097, IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY, IEEE, IECON 2021, Virtual at Toronto, Canada, 13/10/21. https://doi.org/10.1109/IECON48115.2021.9589097

Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles. / van der Heijden, Ties ; Palensky, Peter ; Abraham, Edo.
IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society: Proceedings. IEEE, 2021. 9589097 (IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles

AU - van der Heijden, Ties

AU - Palensky, Peter

AU - Abraham, Edo

N1 - Conference code: 47th

PY - 2021

Y1 - 2021

N2 - In this paper we propose a Quantile Regression Deep Neural Network capable of forecasting multiple quantiles in one model using a combined quantile loss function, and apply it to probabilistically forecast the prices of 8 European Day Ahead Markets. We show that the proposed loss function significantly reduces the quantile crossing problem to (near) 0% in all markets considered, while in some cases simultaneously increasing forecasting performance based on classical point forecast metrics applied to the expected value of the probabilistic forecast. The models are optimized using an automated approach with an elaborate feature- and hyperparameter search space, leading to good model performance in all considered markets.

AB - In this paper we propose a Quantile Regression Deep Neural Network capable of forecasting multiple quantiles in one model using a combined quantile loss function, and apply it to probabilistically forecast the prices of 8 European Day Ahead Markets. We show that the proposed loss function significantly reduces the quantile crossing problem to (near) 0% in all markets considered, while in some cases simultaneously increasing forecasting performance based on classical point forecast metrics applied to the expected value of the probabilistic forecast. The models are optimized using an automated approach with an elaborate feature- and hyperparameter search space, leading to good model performance in all considered markets.

KW - Quantile Regression

KW - Electricity Price Forecasting

KW - Deep Neural Network

KW - Day Ahead Market

KW - Crossing Quantile Problem

UR - http://www.scopus.com/inward/record.url?scp=85119524495&partnerID=8YFLogxK

U2 - 10.1109/IECON48115.2021.9589097

DO - 10.1109/IECON48115.2021.9589097

M3 - Conference contribution

SN - 978-1-6654-0256-9

T3 - IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY

BT - IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society

PB - IEEE

T2 - IECON 2021

Y2 - 13 October 2021 through 16 October 2021

ER -

van der Heijden T , Palensky P , Abraham E. Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles. In IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society: Proceedings. IEEE. 2021. 9589097. (IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY). doi: 10.1109/IECON48115.2021.9589097

Probabilistic DAM price forecasting using a combined Quantile Regression Deep Neural Network with less-crossing quantiles

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this