Fast bootstrap applied to LS-SVM for long term prediction of time series
Amaury Lendasse, Vincent Wertz, Gyula Simon, Michel Verleysen · 2005
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.