[R] Fwd: Extract fixed effects SE from lmer

[Douglas Bates]

I forgot to cc: the list on this reply.

---------- Forwarded message ----------
From: Douglas Bates <dmbates at gmail.com>
Date: Jun 29, 2005 6:28 PM
Subject: Re: [R] Extract fixed effects SE from lmer
To: "La Sorte, Frank A." <lasortef at missouri.edu>


On 6/29/05, La Sorte, Frank A. <lasortef at missouri.edu> wrote:
> Hi,
>
> Does anyone know how to extract fixed effects SE values from generalized linear mixed models estimated using the lmer function in the lme4 library?  I searched attributes and structure with no luck.
>

Try vcov(myLmerModel).  That returns a variance-covariance matrix for
the fixed effects.  The usual manipulations are used to get the
standard errors.  For the binomial and Poisson families you should use
the optional argument useScale=FALSE as these families do not have a
separately estimated scale parameter in the variance function.

In the "show" method for the summary.lmer class the relevant piece of code is

              corF <- as(as(vcov(object, useScale = useScale), "pdmatrix"),
                         "corrmatrix")