CMStatistics 2020: Start Registration
View Submission - CMStatistics
Title: Clustering via copula-based dissimilarity measures Authors:  Pier Giovanni Bissiri - University of Bologna (Italy) [presenting]
Marta Nai Ruscone - Università degli Studi di Genova (Italy)
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.