Gaussian RPMM Split Criterion: Always Split and Record Everything

Share:

Description

Split criterion function: always split, but record everything as you go.

Usage

1
glcSplitCriterionJustRecordEverything(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. This function ALWAYS returns split=TRUE. Useful for gathering information. It is recommended that you set the maxlev argument in the main function to something less than infinity (say, 3 or 4). See glcTree for example of using “glcSplitCriterion...” to control split.

Value

llike1

Just returns llike1

llike2

Just returns llike2

J

Just returns J

weight

Just returns weight

ww

Just returns ww

degFreedom

Degrees-of-freedom for LRT

chiSquareStat

Chi-square statistic

split

TRUE=split the node, FALSE=do not split the node.

See Also

glcSplitCriterionBIC, glcSplitCriterionBICICL, glcSplitCriterionLevelWtdBIC, glcSplitCriterionLRT

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.