GA-SVM based framework for time series forecasting

Nguyen Thi; Gordon Brown Lee; Peter Wheeler; Jim Peterson

doi:10.1109/ICNC.2009.292

GA-SVM based framework for time series forecasting

Nguyen Thi, Gordon Brown Lee, Peter Wheeler, Jim Peterson

Econometrics & Business Statistics

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

6 Citations (Scopus)

Abstract

A framework (hereby named GA-SVM) for time series forecasting was formed by integration of the particular power of Genetic Algorithms (GAs) with the modeling power of the Support Vector Machine (SVM). The proposed system has potential to capture the benefits of both fascinating fields into a single framework. GAs offer high capability in choosing inputs that are relevant and necessary in predicting dependent variables. With these selected inputs, SVM becomes more accurate in modeling the estimation problems. Experiments demonstrated that the integrated GA-SVM approach is superior compared to conventional SVM applications.

Original language	English
Title of host publication	5th International Conference on Natural Computation, ICNC 2009
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	493-498
Number of pages	6
ISBN (Print)	9780769537368
DOIs	https://doi.org/10.1109/ICNC.2009.292
Publication status	Published - 2009
Event	International Conference on Natural Computation 2009 - Tianjian, China Duration: 14 Aug 2009 → 16 Aug 2009 Conference number: 5th https://ieeexplore.ieee.org/xpl/conhome/5362492/proceeding (Proceedings)

Conference

Conference	International Conference on Natural Computation 2009
Abbreviated title	ICNC 2009
Country/Territory	China
City	Tianjian
Period	14/08/09 → 16/08/09
Internet address	https://ieeexplore.ieee.org/xpl/conhome/5362492/proceeding (Proceedings)

Access to Document

10.1109/ICNC.2009.292

Cite this

@inproceedings{ee94923220bc479bbacb02d72da0ea17,

title = "GA-SVM based framework for time series forecasting",

abstract = "A framework (hereby named GA-SVM) for time series forecasting was formed by integration of the particular power of Genetic Algorithms (GAs) with the modeling power of the Support Vector Machine (SVM). The proposed system has potential to capture the benefits of both fascinating fields into a single framework. GAs offer high capability in choosing inputs that are relevant and necessary in predicting dependent variables. With these selected inputs, SVM becomes more accurate in modeling the estimation problems. Experiments demonstrated that the integrated GA-SVM approach is superior compared to conventional SVM applications.",

author = "Nguyen Thi and Lee, {Gordon Brown} and Peter Wheeler and Jim Peterson",

year = "2009",

doi = "10.1109/ICNC.2009.292",

language = "English",

isbn = "9780769537368",

pages = "493--498",

booktitle = "5th International Conference on Natural Computation, ICNC 2009",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "International Conference on Natural Computation 2009, ICNC 2009 ; Conference date: 14-08-2009 Through 16-08-2009",

url = "https://ieeexplore.ieee.org/xpl/conhome/5362492/proceeding",

}

TY - GEN

T1 - GA-SVM based framework for time series forecasting

AU - Thi, Nguyen

AU - Lee, Gordon Brown

AU - Wheeler, Peter

AU - Peterson, Jim

N1 - Conference code: 5th

PY - 2009

Y1 - 2009

N2 - A framework (hereby named GA-SVM) for time series forecasting was formed by integration of the particular power of Genetic Algorithms (GAs) with the modeling power of the Support Vector Machine (SVM). The proposed system has potential to capture the benefits of both fascinating fields into a single framework. GAs offer high capability in choosing inputs that are relevant and necessary in predicting dependent variables. With these selected inputs, SVM becomes more accurate in modeling the estimation problems. Experiments demonstrated that the integrated GA-SVM approach is superior compared to conventional SVM applications.

AB - A framework (hereby named GA-SVM) for time series forecasting was formed by integration of the particular power of Genetic Algorithms (GAs) with the modeling power of the Support Vector Machine (SVM). The proposed system has potential to capture the benefits of both fascinating fields into a single framework. GAs offer high capability in choosing inputs that are relevant and necessary in predicting dependent variables. With these selected inputs, SVM becomes more accurate in modeling the estimation problems. Experiments demonstrated that the integrated GA-SVM approach is superior compared to conventional SVM applications.

UR - http://www.scopus.com/inward/record.url?scp=77950561223&partnerID=8YFLogxK

U2 - 10.1109/ICNC.2009.292

DO - 10.1109/ICNC.2009.292

M3 - Conference Paper

AN - SCOPUS:77950561223

SN - 9780769537368

SP - 493

EP - 498

BT - 5th International Conference on Natural Computation, ICNC 2009

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - International Conference on Natural Computation 2009

Y2 - 14 August 2009 through 16 August 2009

ER -

GA-SVM based framework for time series forecasting

Abstract

Conference

Access to Document

Other files and links

Cite this