bayesclust: An R Package for Testing and Searching for Significant Clusters

This is the publisher’s final pdf. The published article is copyrighted by American Statistical Association and can be found at: http://www.jstatsoft.org/.The detection and determination of clusters has been of special interest among re-\ud searchers from different fields for a long time. In particular, assessing whether the clusters\ud are significant is a question that has been asked by a number of experimenters. In Fuentes\ud and Casella (2009), the authors put forth a new methodology for analyzing clusters. It\ud tests the hypothesis H₀ : = 1 versus H₁ : = k in a Bayesian setting, where denotes\ud the number of clusters in a population. The bayesclust package implements this approach\ud in R. Here we give an overview of the algorithm and a detailed description of the functions\ud available in the package. The routines in bayesclust allow the user to test for the exis-\ud tence of clusters, and then pick out optimal partitionings of the data. We demonstrate\ud the testing procedure with simulated datasets