R: F-test based on the Satterthwaite's approximation for...
calcSatterth
R Documentation
F-test based on the Satterthwaite's approximation for denominator degrees of freedom.
Description
Produces a list with the values for an approximate F-test based on the Satterthwaite's approximation.
Usage
calcSatterth(model, L)
Arguments
model
linear mixed effects model (lmer object).
L
hypothesis contrast matrix or a vector
...
other potential arguments.
Details
F test for the null hypothesis H_0: L β
= 0, where β is a vector of the same length as
fixef(model)
Value
A list with the results from the F test
denom
numeric. Denominator degrees of freedom, calculated with the Satterthwaite's approximation
Fstat
numeric. F statistic
pvalue
numeric. p-value of the corresponding F test
ndf
numeric. Numerator degrees of freedom
Author(s)
Alexandra Kuznetsova, Per Bruun Brockhoff, Rune Haubo Bojesen Christensen
References
Schaalje G.B., McBride J.B., Fellingham G.W. 2002 Adequacy of approximations to distributions of test Statistics in complex mixed linear models
See Also
anova
Examples
## import lme4 package and lmerTest package
library(lmerTest)
## specify lmer model for the sleepstudy data from the lme4 package
m <- lmer(Reaction ~ Days + (1 + Days|Subject), sleepstudy)
L <- cbind(0,1) ## specify contrast vector
calcSatterth(m, L) ## calculate F test
## specify model for the ham data
m.ham <- lmer(Informed.liking ~ Product + (1|Consumer), data = ham)
## specify contrast vector for testing product effect
L <- matrix(0, ncol = 4, nrow = 3)
L[1,2] <- L[2,3] <- L[3,4] <- 1
calcSatterth(m.ham, L)
## by using anova function we get the same result
anova(m.ham)