Wavelet-based relevance vector machines for stock index forecasting

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Relevance vector machine (RVM) is a Beyesian version of the support vector machine, which with a sparse model representation, has appeared as a powerful tool for time series forecasting. RVM has demonstrated better performance over other methods such as neural networks or ARIMA-based models. This paper proposes a wavelet-based RVM model to forecast stock indices. The time series of explanatory variables are decomposed by the wavelet basis, and the extracted time scale features served as inputs of a RVM to perform the nonparametric regression and forecasting. Compared with the traditional GARCH model forecasts, the new method shows superior performance, and reduces the root-mean-squared forecasting errors by nearly one order.

Original languageEnglish
Title of host publicationInternational Joint Conference on Neural Networks 2006, IJCNN '06
Pages603-609
Number of pages7
Publication statusPublished - 2006 Dec 1
EventInternational Joint Conference on Neural Networks 2006, IJCNN '06 - Vancouver, BC, Canada
Duration: 2006 Jul 162006 Jul 21

Other

OtherInternational Joint Conference on Neural Networks 2006, IJCNN '06
CountryCanada
CityVancouver, BC
Period06-07-1606-07-21

Fingerprint

Time series
Support vector machines
Neural networks

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Huang, S-C., & Wu, T-K. (2006). Wavelet-based relevance vector machines for stock index forecasting. In International Joint Conference on Neural Networks 2006, IJCNN '06 (pp. 603-609). [1716149]
Huang, Shian-Chang ; Wu, Tung-Kuang. / Wavelet-based relevance vector machines for stock index forecasting. International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. pp. 603-609
@inproceedings{67df8e5cd22b4e59953466d86628ca10,
title = "Wavelet-based relevance vector machines for stock index forecasting",
abstract = "Relevance vector machine (RVM) is a Beyesian version of the support vector machine, which with a sparse model representation, has appeared as a powerful tool for time series forecasting. RVM has demonstrated better performance over other methods such as neural networks or ARIMA-based models. This paper proposes a wavelet-based RVM model to forecast stock indices. The time series of explanatory variables are decomposed by the wavelet basis, and the extracted time scale features served as inputs of a RVM to perform the nonparametric regression and forecasting. Compared with the traditional GARCH model forecasts, the new method shows superior performance, and reduces the root-mean-squared forecasting errors by nearly one order.",
author = "Shian-Chang Huang and Tung-Kuang Wu",
year = "2006",
month = "12",
day = "1",
language = "English",
isbn = "0780394909",
pages = "603--609",
booktitle = "International Joint Conference on Neural Networks 2006, IJCNN '06",

}

Huang, S-C & Wu, T-K 2006, Wavelet-based relevance vector machines for stock index forecasting. in International Joint Conference on Neural Networks 2006, IJCNN '06., 1716149, pp. 603-609, International Joint Conference on Neural Networks 2006, IJCNN '06, Vancouver, BC, Canada, 06-07-16.

Wavelet-based relevance vector machines for stock index forecasting. / Huang, Shian-Chang; Wu, Tung-Kuang.

International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. p. 603-609 1716149.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Wavelet-based relevance vector machines for stock index forecasting

AU - Huang, Shian-Chang

AU - Wu, Tung-Kuang

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Relevance vector machine (RVM) is a Beyesian version of the support vector machine, which with a sparse model representation, has appeared as a powerful tool for time series forecasting. RVM has demonstrated better performance over other methods such as neural networks or ARIMA-based models. This paper proposes a wavelet-based RVM model to forecast stock indices. The time series of explanatory variables are decomposed by the wavelet basis, and the extracted time scale features served as inputs of a RVM to perform the nonparametric regression and forecasting. Compared with the traditional GARCH model forecasts, the new method shows superior performance, and reduces the root-mean-squared forecasting errors by nearly one order.

AB - Relevance vector machine (RVM) is a Beyesian version of the support vector machine, which with a sparse model representation, has appeared as a powerful tool for time series forecasting. RVM has demonstrated better performance over other methods such as neural networks or ARIMA-based models. This paper proposes a wavelet-based RVM model to forecast stock indices. The time series of explanatory variables are decomposed by the wavelet basis, and the extracted time scale features served as inputs of a RVM to perform the nonparametric regression and forecasting. Compared with the traditional GARCH model forecasts, the new method shows superior performance, and reduces the root-mean-squared forecasting errors by nearly one order.

UR - http://www.scopus.com/inward/record.url?scp=40649091136&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40649091136&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:40649091136

SN - 0780394909

SN - 9780780394902

SP - 603

EP - 609

BT - International Joint Conference on Neural Networks 2006, IJCNN '06

ER -

Huang S-C, Wu T-K. Wavelet-based relevance vector machines for stock index forecasting. In International Joint Conference on Neural Networks 2006, IJCNN '06. 2006. p. 603-609. 1716149