Double SOM for Long-term Time Series Prediction
(2003) WSOM 2003, Workshop on Self-Organizing Maps, Kyushu Institute of Technology — Location: Hibikino (Japan) (11.September.2003)
Files
79-DoubleSOMforlong-termtimeseriesprediction.pdf
Restricted Access - Adobe PDF
- 285.8 KB
Details
- Authors
- Abstract
- Many time series forecasting problems require the estimation of possibly inaccurate, but longterm, 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 longterm 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.
- Affiliations
Université Paris 1 SAMOS-MATISSE Université Jean-Paul Sabatier, Toulouse Laboratoire Statistiques et Probabilités
Citations
Simon, G., Lendasse, A., Cottrell, M., Fort, J.-C., & Verleysen, M. (2003). Double SOM for Long-term Time Series Prediction. Proceedings of WSOM 2003, Workshop on Self-Organizing Maps, p. 340-345. https://hdl.handle.net/2078.5/220834