Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models

Hang T. Nguyen, Ian T. Nabney,

Research output: Contribution to journalArticle

Abstract

This paper presents some forecasting techniques for energy demand and price prediction, one day ahead. These techniques combine wavelet transform (WT) with fixed and adaptive machine learning/time series models (multi-layer perceptron (MLP), radial basis functions, linear regression, or GARCH). To create an adaptive model, we use an extended Kalman filter or particle filter to update the parameters continuously on the test set. The adaptive GARCH model is a new contribution, broadening the applicability of GARCH methods. We empirically compared two approaches of combining the WT with prediction models: multicomponent forecasts and direct forecasts. These techniques are applied to large sets of real data (both stationary and non-stationary) from the UK energy markets, so as to provide comparative results that are statistically stronger than those previously reported. The results showed that the forecasting accuracy is significantly improved by using the WT and adaptive models. The best models on the electricity demand/gas price forecast are the adaptive MLP/GARCH with the multicomponent forecast; their MSEs are 0.02314 and 0.15384 respectively.
Original languageEnglish
Pages (from-to)3674-3685
Number of pages12
JournalEnergy
Volume35
Issue number9
DOIs
Publication statusPublished - Sep 2010

Fingerprint

Wavelet transforms
Electricity
Gases
Multilayer neural networks
Extended Kalman filters
Linear regression
Learning systems
Time series

Bibliographical note

NOTICE: this is the author’s version of a work that was accepted for publication in Energy. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Nguyen, Hang T. and Nabney, Ian T. (2010). Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models. Energy, 35 (9), pp. 3674-3685. DOI 10.1016/j.energy.2010.05.013

Keywords

  • multi-layer perceptron
  • radial basis function
  • GARCH
  • linear regression
  • adaptive models
  • wavelet transform

Cite this

Nguyen, Hang T. ; Nabney, Ian T. / Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models. In: Energy. 2010 ; Vol. 35, No. 9. pp. 3674-3685.
@article{40c4353634d04882b5c4a5d0f1858e41,
title = "Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models",
abstract = "This paper presents some forecasting techniques for energy demand and price prediction, one day ahead. These techniques combine wavelet transform (WT) with fixed and adaptive machine learning/time series models (multi-layer perceptron (MLP), radial basis functions, linear regression, or GARCH). To create an adaptive model, we use an extended Kalman filter or particle filter to update the parameters continuously on the test set. The adaptive GARCH model is a new contribution, broadening the applicability of GARCH methods. We empirically compared two approaches of combining the WT with prediction models: multicomponent forecasts and direct forecasts. These techniques are applied to large sets of real data (both stationary and non-stationary) from the UK energy markets, so as to provide comparative results that are statistically stronger than those previously reported. The results showed that the forecasting accuracy is significantly improved by using the WT and adaptive models. The best models on the electricity demand/gas price forecast are the adaptive MLP/GARCH with the multicomponent forecast; their MSEs are 0.02314 and 0.15384 respectively.",
keywords = "multi-layer perceptron, radial basis function, GARCH, linear regression, adaptive models, wavelet transform",
author = "Nguyen, {Hang T.} and Nabney, {Ian T.}",
note = "NOTICE: this is the author’s version of a work that was accepted for publication in Energy. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Nguyen, Hang T. and Nabney, Ian T. (2010). Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models. Energy, 35 (9), pp. 3674-3685. DOI 10.1016/j.energy.2010.05.013",
year = "2010",
month = "9",
doi = "10.1016/j.energy.2010.05.013",
language = "English",
volume = "35",
pages = "3674--3685",
journal = "Energy",
issn = "0360-5442",
publisher = "Elsevier",
number = "9",

}

Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models. / Nguyen, Hang T.; Nabney, Ian T.

In: Energy, Vol. 35, No. 9, 09.2010, p. 3674-3685.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models

AU - Nguyen, Hang T.

AU - Nabney, Ian T.

N1 - NOTICE: this is the author’s version of a work that was accepted for publication in Energy. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Nguyen, Hang T. and Nabney, Ian T. (2010). Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models. Energy, 35 (9), pp. 3674-3685. DOI 10.1016/j.energy.2010.05.013

PY - 2010/9

Y1 - 2010/9

N2 - This paper presents some forecasting techniques for energy demand and price prediction, one day ahead. These techniques combine wavelet transform (WT) with fixed and adaptive machine learning/time series models (multi-layer perceptron (MLP), radial basis functions, linear regression, or GARCH). To create an adaptive model, we use an extended Kalman filter or particle filter to update the parameters continuously on the test set. The adaptive GARCH model is a new contribution, broadening the applicability of GARCH methods. We empirically compared two approaches of combining the WT with prediction models: multicomponent forecasts and direct forecasts. These techniques are applied to large sets of real data (both stationary and non-stationary) from the UK energy markets, so as to provide comparative results that are statistically stronger than those previously reported. The results showed that the forecasting accuracy is significantly improved by using the WT and adaptive models. The best models on the electricity demand/gas price forecast are the adaptive MLP/GARCH with the multicomponent forecast; their MSEs are 0.02314 and 0.15384 respectively.

AB - This paper presents some forecasting techniques for energy demand and price prediction, one day ahead. These techniques combine wavelet transform (WT) with fixed and adaptive machine learning/time series models (multi-layer perceptron (MLP), radial basis functions, linear regression, or GARCH). To create an adaptive model, we use an extended Kalman filter or particle filter to update the parameters continuously on the test set. The adaptive GARCH model is a new contribution, broadening the applicability of GARCH methods. We empirically compared two approaches of combining the WT with prediction models: multicomponent forecasts and direct forecasts. These techniques are applied to large sets of real data (both stationary and non-stationary) from the UK energy markets, so as to provide comparative results that are statistically stronger than those previously reported. The results showed that the forecasting accuracy is significantly improved by using the WT and adaptive models. The best models on the electricity demand/gas price forecast are the adaptive MLP/GARCH with the multicomponent forecast; their MSEs are 0.02314 and 0.15384 respectively.

KW - multi-layer perceptron

KW - radial basis function

KW - GARCH

KW - linear regression

KW - adaptive models

KW - wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=77955270910&partnerID=8YFLogxK

U2 - 10.1016/j.energy.2010.05.013

DO - 10.1016/j.energy.2010.05.013

M3 - Article

VL - 35

SP - 3674

EP - 3685

JO - Energy

JF - Energy

SN - 0360-5442

IS - 9

ER -