Double SOM for long-term time series prediction - Archive ouverte HAL
Communication Dans Un Congrès Année : 2003

Double SOM for long-term time series prediction

Résumé

Many time series forecasting problems require the estimation of possibly inaccurate, but long¬term, trends, rather than accurate short-term prediction. In this paper, a double use of the Self-Organizing Map algorithm makes it possible to build a model for long¬term prediction, which is proven to be stable. The method uses the information on the structure of the series when available, by predicting blocs instead of scalar values. It is illustrated on real time series for both scalar and bloc predictions.
Fichier principal
Vignette du fichier
WSOM03_-_G._SIMON.pdf (198.96 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00139906 , version 1 (04-04-2007)

Identifiants

  • HAL Id : hal-00139906 , version 1

Citer

Geoffroy Simon, Amaury Lendasse, Marie Cottrell, Jean-Claude Fort, Michel Verleysen. Double SOM for long-term time series prediction. 2003, pp.35-40. ⟨hal-00139906⟩
165 Consultations
178 Téléchargements

Partager

More