sigclust: Statistical Significance of Clustering (sigclust)

In pkimes/sigclust2: sigclust2: Statistical Significance of Clustering

Description

Re-implementation of the Monte Carlo simulation based significance testing procedure described in Liu et al. (2008).

Usage

sigclust(x, n_sim = 100, n_start = 1, icovest = 1,
  bkgd_pca = FALSE, labels = NULL)

Arguments

x	a dataset with n rows and p columns, with observations in rows and features in columns.
n_sim	a numeric value specifying the number of simulations for Monte Carlo testing. (default = 100)
n_start	a numeric value specifying the number of random starts to be used for k-means clustering passed to kmeans. (default = 1)
icovest	an integer (1, 2 or 3) specifying the null covariance estimation method to be used. See null_eigval for more details. (default = 1)
bkgd_pca	a logical value specifying whether to use scaled PCA scores over raw data to estimate the background noise. When FALSE, raw estimate is used; when TRUE, minimum of PCA and raw estimates is used. (default = FALSE)
labels	a n-vector of 1s and 2s specifying cluster labels for testing instead of using kmeans. (default = NULL)

Value

The function returns a sigclust S3-object containing the resulting p-values. The sigclust object has following attributes:

• in_mat: the original data matrix passed to the constructor

Author(s)

Patrick Kimes

References

• Liu Y., Hayes, D. N., Nobel, A., and Marron, J. S. (2008) Statistical significance of clustering for high-dimension, low-sample size data. Journal of the American Statistical Association.

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