R Graphical Manual

Browse All

Last data update: 2014.03.03

R: Beta RPMM Split Criterion: Use ICL-BIC

blcSplitCriterionBICICL

R Documentation

Beta RPMM Split Criterion: Use ICL-BIC

Description

Split criterion function: compare ICL-BICs to determine split (i.e. include entropy term in comparison).

Usage

blcSplitCriterionBICICL(llike1, llike2, weight, ww, J, level)

Arguments

`llike1`	one-class likelihood.
`llike2`	two-class likelihood.
`weight`	weights from RPMM node.
`ww`	“ww” from RPMM node.
`J`	Number of items.
`level`	Node level.

Details

This is a function of the form “glcSplitCriterion...”, which is required to return a list with at least a boolean value split, along with supporting information. See blcTree for example of using “blcSplitCriterion...” to control split.

Value

`bic1`	one-class (weighted) BIC
`bic2`	two-class (weighted) BIC
`entropy`	two-class entropy
`split`	`TRUE`=split the node, `FALSE`=do not split the node.