Using support vector machines for time series prediction

Thissen, Uwe; Brakel, R van; Weijer, A.P. de; Melssen, W.J.; Buydens, L.M.C.

doi:10.1016/s0169-7439(03)00111-4

Cited by 396 publications

(158 citation statements)

References 6 publications

Supporting

Mentioning

153

Contrasting

Unclassified

Order By: Relevance

“…Originally, SVMs have been successfully applied to pattern recognition problems (Burges, 1998;Hsu et al, 2003). However, along with the introduction of Vapnik's ε insensitive loss function, SVMs have been extended to solve nonlinear regression estimation (Gunn, 1998;Smola & Schölkopf, 2004) and time series forecasting (Thissen et al, 2003). It is useful to note that the SVM is finding its way into the water sector (Liong & Sivapragasm, 2002;Bray & Han, 2004;Asefa et al, 2004) and a combination of SVM and evolutionary algorithm called EC-SVM has also been attempted recently (Yu et al, 2004).…”

Section: Introductionmentioning

confidence: 99%

Using support vector machines for long-term discharge prediction

Lin

Cheng

Chau

2006

Hydrological Sciences Journal

518

247

View full text Add to dashboard Cite

Accurate time-and site-specific forecasts of streamflow and reservoir inflow are important in effective hydropower reservoir management and scheduling. Traditionally, autoregressive movingaverage (ARMA) models have been used in modelling water resource time series as a standard representation of stochastic time series. Recently, artificial neural network (ANN) approaches have been proven to be efficient when applied to hydrological prediction. In this paper, the support vector machine (SVM) is presented as a promising method for hydrological prediction. Over-fitting and local optimal solution are unlikely to occur with SVM, which implements the structural risk minimization principle rather than the empirical risk minimization principle. In order to identify appropriate parameters of the SVM prediction model, a shuffled complex evolution algorithm is performed through exponential transformation. The SVM prediction model is tested using the long-term observations of discharges of monthly river flow discharges in the Manwan Hydropower Scheme. Through the comparison of its performance with those of the ARMA and ANN models, it is demonstrated that SVM is a very potential candidate for the prediction of long-term discharges.Key words autoregressive moving-average (ARMA) models; long-term discharge prediction; neural networks; SCE-UA algorithm; support vector machine Utilisation de "support vector machines" pour la prévision de débit à long terme Résumé La gestion et la programmation efficaces d'un barrage hydroélectrique requièrent des prévisions précises, dans le temps et spécifiques pour chaque site, de débit de cours d'eau et de flux entrant dans le réservoir. Les modèles autorégressifs à moyenne mobile (ARMA) sont traditionnellement utilisés pour la modélisation de séries temporelles hydrologiques comme une représentation standard de séries temporelles stochastiques. Récemment des approches par réseaux de neurones artificiels (RNA) se sont révélées être efficaces pour la prévision hydrologique. Dans cet article, une "support vector machine" (SVM) est présentée comme une méthode prometteuse pour la prévision hydrologique. L'approche SVM devrait éviter le sur-apprentisage et les optima locaux, car elle s'appuie sur le principe de minimisation structurelle du risque plutôt que sur le principe de minimisation empirique du risque. Un algorithme "shuffled complex evolution" est utilisé via une transformation exponentielle pour identifier les paramètres appropriés du modèle SVM de prévision. Le modèle SVM de prévision est testé avec les longues séries d'observation des débits mensuels de cours d'eau du complexe hydroélectrique de Manwan. Il apparaît, par comparaison de ses performances avec celles de modèles ARMA et RNA, que l'approche SVM est très pertinente pour la prévision de débits à long terme.

show abstract

Section: Introductionmentioning

confidence: 99%

Using support vector machines for long-term discharge prediction

Lin

Cheng

Chau

2006

Hydrological Sciences Journal

518

247

View full text Add to dashboard Cite

show abstract

“…Some researchers have made some efforts to overcome these by applying an adapted form called weighted LS-SVM (Thissen et al, 2003).…”

Section: Discussionmentioning

confidence: 99%

Application of least squares vector machines in modelling water vapor and carbon dioxide fluxes over a cropland

Qin

et al. 2005

J Zheijang Univ Sci B

View full text Add to dashboard Cite

Least squares support vector machines (LS-SVMs), a nonlinear kemel based machine was introduced to investigate the prospects of application of this approach in modelling water vapor and carbon dioxide fluxes above a summer maize field using the dataset obtained in the North China Plain with eddy covariance technique. The performances of the LS-SVMs were compared to the corresponding models obtained with radial basis function (RBF) neural networks. The results indicated the trained LS-SVMs with a radial basis function kernel had satisfactory performance in modelling surface fluxes; its excellent approximation and generalization property shed new light on the study on complex processes in ecosystem.

show abstract

“…The scale of the simulation experiment is small in both cases. Thissen et al (2003) examine one long time series from the ARMA family, while Zhang (2001) examine 8 stochastic processes from the ARMA family and 30 simulated time series for each stochastic process. The forecasting methods are ARMA models, NN and SVM in the former study and ARMA models and NN in the latter study, while Makridakis and Hibon (1987), Makridakis and Hibon (2000) and Ahmed et al (2010) do not focus their comparisons on the stochastic-ML dipole.…”

Section: The Broader Perspectivementioning

confidence: 99%

Comparison of Stochastic and Machine Learning Methods for Multi-Step Ahead Forecasting of Hydrological Processes

Papacharalampous¹,

Tyralis²,

Koutsoyiannis³

2017

Preprint

View full text Add to dashboard Cite

Abstract:We perform an extensive comparison between 11 stochastic to 9 machine learning methods regarding their multi-step ahead forecasting properties by conducting 12 large-scale computational experiments. Each of these experiments uses 2 000 time series generated by linear stationary stochastic processes. We conduct each simulation experiment twice; the first time using time series of 110 values and the second time using time series of 310 values. Additionally, we conduct 92 real-world case studies using mean monthly time series of streamflow and particularly focus on one of them to reinforce the findings and highlight important facts. We quantify the performance of the methods using 18 metrics. The results indicate that the machine learning methods do not differ dramatically from the stochastic, while none of the methods under comparison is uniformly better or worse than the rest. However, there are methods that are regularly better or worse than others according to specific metrics.

show abstract

Using support vector machines for time series prediction

Cited by 396 publications

References 6 publications

Using support vector machines for long-term discharge prediction

Using support vector machines for long-term discharge prediction

Application of least squares vector machines in modelling water vapor and carbon dioxide fluxes over a cropland

Comparison of Stochastic and Machine Learning Methods for Multi-Step Ahead Forecasting of Hydrological Processes

Contact Info

Product

Resources

About