Title :
Fast bootstrap applied to LS-SVM for long term prediction of time series
Author :
Lendasse, Amaury ; Wertz, Vincent ; Simon, Geoffroy ; Verleysen, Michel
Author_Institution :
CIS, HUT, Finland
Abstract :
Time series forecasting is usually limited to one-step ahead prediction. This goal is extended here to longer-term prediction, obtained using the least-square support vector machines model. The influence of the model parameters is observed when the time horizon of the prediction is increased and for various prediction methods. The model selection to optimize the design parameters is performed using the fast bootstrap methodology introduced in previous works.
Keywords :
least squares approximations; support vector machines; time series; LS-SVM; fast bootstrap; least-square support vector machines model; time series forecasting; time series prediction; Computational Intelligence Society; Design optimization; Finance; Floods; Lagrangian functions; Load forecasting; Prediction methods; Predictive models; Rivers; Support vector machines;
Conference_Titel :
Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on
Print_ISBN :
0-7803-8359-1
DOI :
10.1109/IJCNN.2004.1380003
