logo-printable
Ricerca

In hierarchical cluster analysis by copula approach


Ricerca
In hierarchical cluster analysis by copula approach

We propose a new measure to evaluate the dissimilarity between rankings in hierarchical cluster analysis to segment subjects expressing their preferences by rankings. The proposed index builds upon the Spearman grade correlation

coefficient on a transformation of the ordinal variables that describes the rankings of the subjects, calculated by the copula function. In particular, in using the copula functions with tail dependence we employ an index suitable for emphasizing the agreement on top ranks, when the top ranks are considered more important than the lower ones. We evaluate the performance of our proposal by an example on selected rankings, showing that the resulting groups contain subjects whose preferences are more similar on the most important, or top, ranks.
Pubblicato il 31 Maggio 2016
Informazioni utili
Piazza Soldini, 5 - Castellanza (VA)
Email: comunicazione@liuc.it |T. +39 0331 572111
ISCRIZIONI
QR Code
Follow us!
logo-liuc
logo-confindustria
Newsletter
Vuoi rimanere aggiornato su eventi e news LIUC?

logo-liuc
logo-confindustria
© UNIVERSITA' CARLO CATTANEO - LIUC | C.SO MATTEOTTI, 22 - 21053 CASTELLANZA (VA)
CODICE FISCALE E PARTITA IVA 02015300128