A theoretical framework for clustering data is presented according to the dissimilarity behaviour as measured via a suitable copula-based coefficient and study its main properties. The coefficients are defined in terms of copulas, which may or may not be Gaussian. Applications to real data are used to illustrate the usefulness and importance of our proposal.
Clustering via copula-based dissimilarity measures
Nai Ruscone, Marta;
2020-01-01
Abstract
A theoretical framework for clustering data is presented according to the dissimilarity behaviour as measured via a suitable copula-based coefficient and study its main properties. The coefficients are defined in terms of copulas, which may or may not be Gaussian. Applications to real data are used to illustrate the usefulness and importance of our proposal.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
1222020.pdf
accesso chiuso
Tipologia:
Documento in versione editoriale
Dimensione
702.12 kB
Formato
Adobe PDF
|
702.12 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
