Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 57 - Nonparametric Modeling I
Type: Contributed
Date/Time: Sunday, August 8, 2021 : 3:30 PM to 5:20 PM
Sponsor: Section on Nonparametric Statistics
Abstract #318221
Title: Permutation tests for cluster analysis
Author(s): Riccardo Ceccato* and Rosa Arboretti and Luigi Salmaso and Luca Pegoraro and Marta Disegna
Companies: University of Padova and University of Padova and University of Padova and University of Padova and Bournemouth University
Keywords: Cluster analysis; Permutation test

Cluster analysis is a powerful, versatile tool used in several fields to classify objects according to a set of observed characteristics. This analysis commonly involves some critical decisions, such as choosing the number of clusters to consider. To help practitioners undertake this specific challenge, we propose a permutation-based approach which makes it possible to compute a ranking of $C$ different partitions into $K_c$, $c=1,\ldots,C$ clusters. In particular, this procedure avoids choosing a single clustering quality index for the choice of optimal number of clusters, and bases the decision on multiple indices. A case study presenting an example of successful application of the aforementioned approach is considered.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program