Comparison of ARIMA and ANN models used in electricity price forecasting for power market

Research output: Contribution to journalArticle

Abstract

In power market, electricity price forecasting provides significant information which can help the electricity market participants to prepare corresponding bidding strategies to maximize their profits. This paper introduces the models of autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) which are applied to the price forecasts for up to 3 steps ahead in the UK electricity market. The half hourly data of historical prices are obtained from UK Reference Price Data from March 22nd to July 14th 2010 and the predictions are derived from a sliding training window with a length of 8 weeks. The ARIMA with various AR and MA orders and the ANN with different numbers of delays and neurons have been established and compared in terms of the root mean square errors (RMSEs) of price forecasts. The experimental results illustrate that the ARIMA (4,1,2) model gives greater improvement over persistence than the ANN (20 neurons, 4 delays) model.
LanguageEnglish
Pages120-126
Number of pages7
JournalEnergy and Power Engineering
Volume9
Issue number4B
DOIs
Publication statusPublished - 6 Apr 2017
Event9th Asia-Pacific Power and Energy Engineering Conference - Chengdu, China
Duration: 15 Apr 201717 Apr 2017

Fingerprint

Electricity
Neural networks
Neurons
Mean square error
Profitability
Power markets

Keywords

  • electricity markets
  • electricity prices
  • ARIMA models
  • ANN models
  • short-term forecasting

Cite this

@article{75863fe61da24347ac1a159b659a1e0f,
title = "Comparison of ARIMA and ANN models used in electricity price forecasting for power market",
abstract = "In power market, electricity price forecasting provides significant information which can help the electricity market participants to prepare corresponding bidding strategies to maximize their profits. This paper introduces the models of autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) which are applied to the price forecasts for up to 3 steps ahead in the UK electricity market. The half hourly data of historical prices are obtained from UK Reference Price Data from March 22nd to July 14th 2010 and the predictions are derived from a sliding training window with a length of 8 weeks. The ARIMA with various AR and MA orders and the ANN with different numbers of delays and neurons have been established and compared in terms of the root mean square errors (RMSEs) of price forecasts. The experimental results illustrate that the ARIMA (4,1,2) model gives greater improvement over persistence than the ANN (20 neurons, 4 delays) model.",
keywords = "electricity markets, electricity prices, ARIMA models, ANN models, short-term forecasting",
author = "Gao Gao and Kwoklun Lo and Fulin Fan",
year = "2017",
month = "4",
day = "6",
doi = "10.4236/epe.2017.94B015",
language = "English",
volume = "9",
pages = "120--126",
journal = "Energy and Power Engineering",
issn = "1949-243X",
number = "4B",

}

Comparison of ARIMA and ANN models used in electricity price forecasting for power market. / Gao, Gao; Lo, Kwoklun; Fan, Fulin.

In: Energy and Power Engineering, Vol. 9, No. 4B, 06.04.2017, p. 120-126.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Comparison of ARIMA and ANN models used in electricity price forecasting for power market

AU - Gao, Gao

AU - Lo, Kwoklun

AU - Fan, Fulin

PY - 2017/4/6

Y1 - 2017/4/6

N2 - In power market, electricity price forecasting provides significant information which can help the electricity market participants to prepare corresponding bidding strategies to maximize their profits. This paper introduces the models of autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) which are applied to the price forecasts for up to 3 steps ahead in the UK electricity market. The half hourly data of historical prices are obtained from UK Reference Price Data from March 22nd to July 14th 2010 and the predictions are derived from a sliding training window with a length of 8 weeks. The ARIMA with various AR and MA orders and the ANN with different numbers of delays and neurons have been established and compared in terms of the root mean square errors (RMSEs) of price forecasts. The experimental results illustrate that the ARIMA (4,1,2) model gives greater improvement over persistence than the ANN (20 neurons, 4 delays) model.

AB - In power market, electricity price forecasting provides significant information which can help the electricity market participants to prepare corresponding bidding strategies to maximize their profits. This paper introduces the models of autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) which are applied to the price forecasts for up to 3 steps ahead in the UK electricity market. The half hourly data of historical prices are obtained from UK Reference Price Data from March 22nd to July 14th 2010 and the predictions are derived from a sliding training window with a length of 8 weeks. The ARIMA with various AR and MA orders and the ANN with different numbers of delays and neurons have been established and compared in terms of the root mean square errors (RMSEs) of price forecasts. The experimental results illustrate that the ARIMA (4,1,2) model gives greater improvement over persistence than the ANN (20 neurons, 4 delays) model.

KW - electricity markets

KW - electricity prices

KW - ARIMA models

KW - ANN models

KW - short-term forecasting

UR - http://www.scirp.org/Journal/PaperInformation.aspx?PaperID=75241

U2 - 10.4236/epe.2017.94B015

DO - 10.4236/epe.2017.94B015

M3 - Article

VL - 9

SP - 120

EP - 126

JO - Energy and Power Engineering

T2 - Energy and Power Engineering

JF - Energy and Power Engineering

SN - 1949-243X

IS - 4B

ER -