Displaying 1 result from an estimated 1 matches for "randomeffects".
Did you mean:
random_effects
2006 Jul 20
1
Loss of numerical precision from conversion to list ?
...e~age,random=~1|Subject,data=Orthodont)
# summary(m1)
# ...
# Random effects:
# Formula: ~1 | Subject
# (Intercept) Residual
# StdDev: 2.114724 1.431592 -> lambda.REML=2.114^2/1.431^2 = 2.182382
#DesignMatrix for fixed Effects
X<-cbind(rep(1,108),Orthodont$age)
#DesignMatrix of RandomEffects
Z<-matrix(data=c(rep(1,4),rep(0,108)),nrow=108,ncol=27)
#Corr(RanEf)^0.5 = 27 x 27 Identity, since RandomIntercepts are independent
sqrt.Sigma<-diag(27)
K<-27 #number of subjects/ random intercepts
n<-nrow(X)
p<-ncol(X)
lambda0 <- 2.182382 #actually not a sensible choice as Nu...