Time-series analysis with a hybrid Box-Jenkins ARIMA and neural network model

Dilli R. Aryal; Yao Wu Wang

Time-series analysis with a hybrid Box-Jenkins ARIMA and neural network model

Dilli R. Aryal
, Yao Wu Wang

School of Management, Harbin Institute of Technology

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

9 Scopus citations

Abstract

In this study, a hybrid ARIMA and neural networks model to time series forecasting is proposed. The basic idea behind the model combination is to use each model's unique features to capture different patterns in the data. With three real data sets, empirical results evidently show that the hybrid model outperforms ARIMA and ANN model being used in isolation in terms of forecasting accuracy.

Original language	English
Pages (from-to)	413-421
Number of pages	9
Journal	Journal of Harbin Institute of Technology (New Series)
Volume	11
Issue number	4
State	Published - Aug 2004
Externally published	Yes

Keywords

ARIMA
Artificial neural networks
Box-Jenkins methodology
Hybrid model
Time series analysis

Cite this

@article{00d360d4787547a2af3254a938cac7df,

title = "Time-series analysis with a hybrid Box-Jenkins ARIMA and neural network model",

abstract = "In this study, a hybrid ARIMA and neural networks model to time series forecasting is proposed. The basic idea behind the model combination is to use each model's unique features to capture different patterns in the data. With three real data sets, empirical results evidently show that the hybrid model outperforms ARIMA and ANN model being used in isolation in terms of forecasting accuracy.",

keywords = "ARIMA, Artificial neural networks, Box-Jenkins methodology, Hybrid model, Time series analysis",

author = "Aryal, \{Dilli R.\} and Wang, \{Yao Wu\}",

year = "2004",

month = aug,

language = "英语",

volume = "11",

pages = "413--421",

journal = "Journal of Harbin Institute of Technology (New Series)",

issn = "1005-9113",

publisher = "Harbin Institute of Technology",

number = "4",

}

TY - JOUR

T1 - Time-series analysis with a hybrid Box-Jenkins ARIMA and neural network model

AU - Aryal, Dilli R.

AU - Wang, Yao Wu

PY - 2004/8

Y1 - 2004/8

N2 - In this study, a hybrid ARIMA and neural networks model to time series forecasting is proposed. The basic idea behind the model combination is to use each model's unique features to capture different patterns in the data. With three real data sets, empirical results evidently show that the hybrid model outperforms ARIMA and ANN model being used in isolation in terms of forecasting accuracy.

AB - In this study, a hybrid ARIMA and neural networks model to time series forecasting is proposed. The basic idea behind the model combination is to use each model's unique features to capture different patterns in the data. With three real data sets, empirical results evidently show that the hybrid model outperforms ARIMA and ANN model being used in isolation in terms of forecasting accuracy.

KW - ARIMA

KW - Artificial neural networks

KW - Box-Jenkins methodology

KW - Hybrid model

KW - Time series analysis

UR - https://www.scopus.com/pages/publications/10044297317

M3 - 文章

AN - SCOPUS:10044297317

SN - 1005-9113

VL - 11

SP - 413

EP - 421

JO - Journal of Harbin Institute of Technology (New Series)

JF - Journal of Harbin Institute of Technology (New Series)

IS - 4

ER -

Time-series analysis with a hybrid Box-Jenkins ARIMA and neural network model

Abstract

Keywords

Other files and links

Fingerprint

Cite this