Forecasting performance of LS-SVM for nonlinear hydrological time series

Seok Hwan Hwang, Dae Heon Ham, Joong Hoon Kim

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

This paper presents a Least-Square Support Vector Machine (LS-SVM) approach for forecasting nonlinear hydrological time series. LS-SVM is a machine-learning algorithm firmly based on the statistical learning theory. The objective of this paper is to examine the feasibility of using LS-SVM in the forecasting of nonlinear hydrological time series by comparing it with a statistical method such as Multiple Linear Regression (MLR) and a heuristic method such as a Neural Network using Back-Propagation (NNBP). And the performance of prediction model is also dependent on the degrees of linearity (or persistency) of data, not only on the performance of model itself. Thus, we would clearly verify that prediction performance of three models according to linear extent using daily water demand and daily inflow of dam data. In the experimental results, LS-SVM showed superior forecasting accuracies and performances to those of MLR and NNBP and LS-SVM demonstrated better forecasting efficiency in nonlinear hydrological time series using Relative Correlation Coefficient (RCC) which is a relative measure of forecasting efficiency with different persistency.

Original languageEnglish
Pages (from-to)870-882
Number of pages13
JournalKSCE Journal of Civil Engineering
Volume16
Issue number5
DOIs
Publication statusPublished - 2012 Jul 1

Fingerprint

Support vector machines
Time series
Backpropagation
Linear regression
Neural networks
Heuristic methods
Learning algorithms
Dams
Learning systems
Statistical methods
Water

Keywords

  • forecasting
  • forecasting performance
  • support vector machine

ASJC Scopus subject areas

  • Civil and Structural Engineering

Cite this

Forecasting performance of LS-SVM for nonlinear hydrological time series. / Hwang, Seok Hwan; Ham, Dae Heon; Kim, Joong Hoon.

In: KSCE Journal of Civil Engineering, Vol. 16, No. 5, 01.07.2012, p. 870-882.

Research output: Contribution to journalArticle

@article{3b3d6ddfbd994288b41c7576822fb1d1,
title = "Forecasting performance of LS-SVM for nonlinear hydrological time series",
abstract = "This paper presents a Least-Square Support Vector Machine (LS-SVM) approach for forecasting nonlinear hydrological time series. LS-SVM is a machine-learning algorithm firmly based on the statistical learning theory. The objective of this paper is to examine the feasibility of using LS-SVM in the forecasting of nonlinear hydrological time series by comparing it with a statistical method such as Multiple Linear Regression (MLR) and a heuristic method such as a Neural Network using Back-Propagation (NNBP). And the performance of prediction model is also dependent on the degrees of linearity (or persistency) of data, not only on the performance of model itself. Thus, we would clearly verify that prediction performance of three models according to linear extent using daily water demand and daily inflow of dam data. In the experimental results, LS-SVM showed superior forecasting accuracies and performances to those of MLR and NNBP and LS-SVM demonstrated better forecasting efficiency in nonlinear hydrological time series using Relative Correlation Coefficient (RCC) which is a relative measure of forecasting efficiency with different persistency.",
keywords = "forecasting, forecasting performance, support vector machine",
author = "Hwang, {Seok Hwan} and Ham, {Dae Heon} and Kim, {Joong Hoon}",
year = "2012",
month = "7",
day = "1",
doi = "10.1007/s12205-012-1519-3",
language = "English",
volume = "16",
pages = "870--882",
journal = "KSCE Journal of Civil Engineering",
issn = "1226-7988",
publisher = "Korean Society of Civil Engineers",
number = "5",

}

TY - JOUR

T1 - Forecasting performance of LS-SVM for nonlinear hydrological time series

AU - Hwang, Seok Hwan

AU - Ham, Dae Heon

AU - Kim, Joong Hoon

PY - 2012/7/1

Y1 - 2012/7/1

N2 - This paper presents a Least-Square Support Vector Machine (LS-SVM) approach for forecasting nonlinear hydrological time series. LS-SVM is a machine-learning algorithm firmly based on the statistical learning theory. The objective of this paper is to examine the feasibility of using LS-SVM in the forecasting of nonlinear hydrological time series by comparing it with a statistical method such as Multiple Linear Regression (MLR) and a heuristic method such as a Neural Network using Back-Propagation (NNBP). And the performance of prediction model is also dependent on the degrees of linearity (or persistency) of data, not only on the performance of model itself. Thus, we would clearly verify that prediction performance of three models according to linear extent using daily water demand and daily inflow of dam data. In the experimental results, LS-SVM showed superior forecasting accuracies and performances to those of MLR and NNBP and LS-SVM demonstrated better forecasting efficiency in nonlinear hydrological time series using Relative Correlation Coefficient (RCC) which is a relative measure of forecasting efficiency with different persistency.

AB - This paper presents a Least-Square Support Vector Machine (LS-SVM) approach for forecasting nonlinear hydrological time series. LS-SVM is a machine-learning algorithm firmly based on the statistical learning theory. The objective of this paper is to examine the feasibility of using LS-SVM in the forecasting of nonlinear hydrological time series by comparing it with a statistical method such as Multiple Linear Regression (MLR) and a heuristic method such as a Neural Network using Back-Propagation (NNBP). And the performance of prediction model is also dependent on the degrees of linearity (or persistency) of data, not only on the performance of model itself. Thus, we would clearly verify that prediction performance of three models according to linear extent using daily water demand and daily inflow of dam data. In the experimental results, LS-SVM showed superior forecasting accuracies and performances to those of MLR and NNBP and LS-SVM demonstrated better forecasting efficiency in nonlinear hydrological time series using Relative Correlation Coefficient (RCC) which is a relative measure of forecasting efficiency with different persistency.

KW - forecasting

KW - forecasting performance

KW - support vector machine

UR - http://www.scopus.com/inward/record.url?scp=84863306895&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863306895&partnerID=8YFLogxK

U2 - 10.1007/s12205-012-1519-3

DO - 10.1007/s12205-012-1519-3

M3 - Article

AN - SCOPUS:84863306895

VL - 16

SP - 870

EP - 882

JO - KSCE Journal of Civil Engineering

JF - KSCE Journal of Civil Engineering

SN - 1226-7988

IS - 5

ER -