Skip to content

lmer confint() bootstrap doesn't match fitted values when weighted

2 messages · Rodrigo Travitzki, Ben Bolker

Original
Rodrigo Travitzki
# 
Dear masters,
I found a strange behaviour in confint() bootstrap methods, maybe 
something with my data, I don't know.

My data is weighted and is something related to it, as you can see in 
code below.

Best,
Rodrigo


## run
## strange behaviour of confint()
#
# R version: 3.3.1
# lme4 version: 1.1-12

require(lme4)

dd=read.csv2(file("http://topicostropicais.net/bau/data.csv"))

m1<-lmer(NOTA~(1|ID_ESCOLA)+(1|ID_TURMA),data=dd,weights=PESO)

## pre conditions seems to be OK
plot(m1)
qqnorm(scale(resid(m1)),ylab="Residual quantiles",col="orange")
qqline(scale(resid(m1)),col="blue")

## reproducing error
VarCorr(m1) # Residual = 46.5197
# with "boot", sigma intervals doesn't match the estimated value 46.5197
confint(m1,method="boot",boot.type="norm")
confint(m1,method="boot",boot.type="basic")
confint(m1,method="boot",boot.type="perc")

## cause seems to be 'weigths'
m2<-lmer(NOTA~(1|ID_ESCOLA)+(1|ID_TURMA),data=dd)
VarCorr(m2)
confint(m2,method="boot")
confint(m2,method="profile")
Ben Bolker
# 
I'm guessing this is a bug.  Weights have to be specially handled and
we probably missed a step here.  Posted as
https://github.com/lme4/lme4/issues/386 ...

  thanks
    Ben Bolker
On 16-06-29 09:13 PM, Rodrigo Travitzki wrote: