sigclust2-package: Statistical Significance for Clustering

Description Author(s)


This package implements statistical significance of clustering procedures based on the assumption that a cluster may be well approximated as a single Gaussian distribution. The procedure is particularly well-suited for the high-dimension low-sample size setting when the number of dimensions greatly outnumbers the number of observations. The package builds substantially on the sigclust package currently available through CRAN to include additional options for null Gaussian estimation based on scaled PCA scores, and more substantially, to test for significance in hierarchical clustering.


Patrick Kimes

pkimes/sigclust2 documentation built on May 25, 2019, 8:20 a.m.