Fast Bootstrap applied to LS-SVM for long Term Prediction of Time Series
(2004) IJCNN′04, 2004 International Joint Conference on Neural Networks — Location: Budapest (Hungary) (25.July.2004)
Files
95-FastBootstrapappliedtoLS-SVMforLongTermPredictionofTimeSeries.pdf
Restricted Access - Adobe PDF
- 330.48 KB
Details
- Authors
- 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.
- Affiliations
Helsinky University of Technology Neural Networks Research Center
Citations
Lendasse, A., Wertz, V., Simon, G., & Verleysen, M. (2004). Fast Bootstrap applied to LS-SVM for long Term Prediction of Time Series. Proceedings of IJCNN 2004, International Joint Conference on Neural Networks, p. 705-710. https://hdl.handle.net/2078.5/225868