KenwardRoger primitive function: Ftest and degrees of freedom based on Kenward-Roger approximation
Description
An approximation Ftest based on the Kenward-Roger approximation fo the
residuals degress of freedom and the scale F-statistic.
Usage
.KRmodcompPrimitive(largeModel,L,beta0)
Arguments
largeModel
A lmer model fitted with REML=TRUE
L
A restriction matrix L
beta0
A number of a vector of the beta of the hypothesis,
e.g. L beta=beta0. beta0=0 if modelSmall is a model not a
restriction matrix.
Details
L is a restriction matrix, such that modelLarge is a lmer-model with fixed
effect model matrix X and fixed effects parameters beta, then
{alpha| L beta=0} defines a submodel model0 of modelLarge with
in
