Dear R users
I encounter a problem when i perform a generalized linear mixed model (binary
data) with the lmer function (package lme4)
with R 2.1.0 on windows XP and the latest version of package lme4 (0.96-1)
and matrix (0.96-2)
both options PQL and Laplace for the method argument in lmer
i am aware of this but i don't know why.
On Thu, 16 Jun 2005 10:31:25 +0200
Emmanuel Tillard [EMAIL PROTECTED] wrote:
Dear R users
I encounter a problem when i perform a generalized linear mixed model (binary
data) with the lmer function (package lme4)
with R 2.1.0 on windows XP and the
It's a dumb mistake in the code. I missed a factor of 2 when
converting from deviance to log-likelihood in one place. I'll upload a
new version of the Matrix package. (It happens that all of the code
for lmer is in the Matrix package now so there is no need to update
the lme4 package.)
On