lrpCV {longRPart}R Documentation

Performs cross-validation on the longRPart model

Description

This functions implements v-fold cross validation on the longRPart trees. IT IS CURRENTLY UNDER DEVELOPMENT AND SHOULD NOT BE USED FOR ANALYSIS.

Usage

lrpCV(model)

Arguments

model model produced by longRPart

Details

Using the xval in the the rpart control object, this function performs v-fold cross-validation. The object returned is deviance between the subject and the predicted values at the node it was navigated to as a test case. The mean of that vector would be the average deviances.

Author(s)

Mohamed Abdolell <mo@dal.ca> and Sam Stewart

References

Abdolell et al. Binary partitioning for continuous longitudinal data: categorizing a prognostic variable. Statistics in Medicine (2002) vol. 21 pp. 3395-3409


[Package longRPart version 1.0 Index]