Clustering multivariate functional data
Résumé
Model-based clustering is considered for Gaussian multivariate functional data as an extension of the univariate functional setting. Principal components analysis is introduced and used to define an approximation of the notion of density for multivariate functional data. An EM like algorithm is proposed to estimate the parameters of the reduced model. Application on climatology data illustrates the method.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...