Autori
Paganoni, Anna Maria
Ghiglietti, Andrea
Leva, Francesca
Martino, Andrea

Titolo
A k-means procedure based on a Mahalanobis type distance for clustering multivariate functional data
Periodico
Statistical methods & applications : Journal of the Italian Statistical Society
Anno: 2019 - Volume: 28 - Fascicolo: 2 - Pagina iniziale: 301 - Pagina finale: 322

This paper proposes a clustering procedure for samples of multivariate functions in (L2(I))J, with J≥1. This method is based on a k-means algorithm in which the distance between the curves is measured with a metric that generalizes the Mahalanobis distance in Hilbert spaces, considering the correlation and the variability along all the components of the functional data. The proposed procedure has been studied in simulation and compared with the k-means based on other distances typically adopted for clustering multivariate functional data. In these simulations, it is shown that the k-means algorithm with the generalized Mahalanobis distance provides the best clustering performances, both in terms of mean and standard deviation of the number of misclassified curves. Finally, the proposed method has been applied to two case studies, concerning ECG signals and growth curves, where the results obtained in simulation are confirmed and strengthened.



SICI: 1618-2510(2019)28:2<301:AKPBOA>2.0.ZU;2-Y

Esportazione dati in Refworks (solo per utenti abilitati)

Biblioteche ACNP che possiedono il periodico