Please use this identifier to cite or link to this item: http://arl.liuc.it/dspace/handle/2468/5933
Title: Dissimilarity measure for ranking data via mixture of copulae
Authors: Bonanomi, Andrea
Nai Ruscone, Marta
Osmetti, Silvia Angela
Issue Date: 2018
Publisher: FedOAPress
Bibliographic citation: Bonanomi Andrea, Nai Ruscone Marta, Osmetti Silvia Angela (2018), Dissimilarity measure for ranking data via mixture of copulae. In: Capecchi Stefania, Di Iorio Francesca, Simone Rosaria, ed., ASMOD 2018: proceedings of the advanced statistical modelling for ordinal data conference: Naples, 24-26 October 2018. (Scuola di scienze umane e sociali. Quaderni, 11). Napoli: FedOAPress, p. 53-59. ISBN 978-88-6887-042-3.
Abstract: We propose a new dissimilarity measure for ranking data by using a mixture of copula functions. This measure evaluates the dissimilarity between subjects expressing their preferences by rankings in order to classify them by a hierarchical cluster analysis. The proposed measure is based on the Spearman’s grade correlation coefficient on a transformation, operated by the copula, of the rank denoting the level of the importance assigned by subjects in the classification process. The mixtures of copulae are a flexible way to model different types of dependence structures in the data and to consider different situations in the classification process. The advantage by using mixtures of copulae with lower and upper tail dependence is that we can emphasize the agreement on extreme ranks, when extreme ranks are considered more important. An example on simulated data illustrates our proposal.
URI: http://arl.liuc.it/dspace/handle/2468/5933
Journal/Book: ASMOD 2018: proceedings of the advanced statistical modelling for ordinal data conference: Naples, 24-26 October 2018
ISBN: 978-88-6887-042-3
Appears in Collections:Contributo in atti di convegno

Files in This Item:
File Description SizeFormat 
5933.pdf
  Restricted Access
639,92 kBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.