cluster.BIC: BIC for subspace clustering

Description Usage Arguments Value

View source: R/auxiliary.functions.R

Description

Computes the value of BIC criterion for given data set and partition. In each cluster we assume that variables are spanned by few factors. Considering maximum likelihood we get that those factors are in fact principal components. Noise sigma can be computed jointly for all clusters (default), seperately for each cluster or be specified as input.

Usage

1
2
cluster.BIC(X, segmentation, max.dim, numb.clusters, sigma = NULL,
  estimateJointly = TRUE)

Arguments

X

a matrix with only continuous variables

segmentation

a vector, segmentation for which likelihood is computed. Clusters numbers should be from range [1, numb.clusters]

max.dim

an integer, maximum dimension of subspace. Number of principal components that span each subspace.

numb.clusters

an integer, number of clusters

sigma

a numeric, (default is NULL) value of sigma provided by the user

estimateJointly

a boolean, (default value is TRUE) indicating if sigma should be estimated jointly for all clusters

Value

BIC value of BIC criterion


psobczyk/public_varclust documentation built on May 24, 2017, 12:20 p.m.