RPMM (version 1.20)

blcSplit: Beta Latent Class Splitter

Description

Splits a data set into two via a beta mixture model

Usage

blcSplit(x, initFunctions, weight = NULL, index = NULL, level = NULL,  wthresh = 1e-09, verbose = TRUE, nthresh = 5,  splitCriterion = NULL)

Arguments

x

Data matrix (n x j) on which to perform clustering

initFunctions

List of functions of type “blcInitialize...” for initializing latent class model. See blcInitializeFanny for an example of arguments and return values.

weight

Weight corresponding to the indices passed (see index). Defaults to 1 for all indices

index

Row indices of data matrix to include. Defaults to all (1 to n).

level

Current level.

wthresh

Weight threshold for filtering data to children. Indices having weight less than this value will not be passed to children nodes.

verbose

Level of verbosity. Default=2 (too much). 0 for quiet.

nthresh

Total weight in node required for node to be a candidate for splitting. Nodes with weight less than this value will never split.

splitCriterion

Function of type “blcSplitCriterion...” for determining whether split should occur. See blcSplitCriterionBIC for an example of arguments and return values. Default behavior is blcSplitCriterionBIC (though the function is bypassed by internal calculations for some modest computational efficiency gains).

Value

A list of objects representing split.

Details

Should not be called by user.