[R-sig-ME] Fwd: cross-classified random effects model R code for empirical bayes

[Douglas Bates]

I forgot to cc: the list on this reply.


---------- Forwarded message ----------
From: Douglas Bates <bates at stat.wisc.edu>
Date: Thu, Sep 27, 2012 at 3:54 PM
Subject: Re: cross-classified random effects model R code for empirical bayes
To: Webster Kasongo <kasongster at yahoo.com>


It is better to send questions like this to the
R-SIG-Mixed-Models at R-project.org mailing list (see
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models for more
information) than to me directly.  Several of those who read that list
can respond to you and often do so much more quickly than I am able
to.

On Thu, Sep 27, 2012 at 3:38 PM, Webster Kasongo <kasongster at yahoo.com> wrote:
> Dear Dr. Bates
> I am womdering whether there is a way of specifying the parameter estimation
> method for empirical bayes method
> e.g.,
> summary(fit.T2 <-lmer(wordsum ~ race+GENDER+ age+ I(age^2)+ educ+
> (1|PERIOD)+(1|COHORT), data=GSSBCFINAL, REML=FALSE))
>
> In the above code, estimation method is ML.

> How can one estimate emprical
> bayesian method?

A facetious answer would be "learn to program in R".  :-)

I don't know of any R packages that provide empirical Bayes estimates.
 In fact, I'm not sure that the name "empirical Bayes" is sufficient
to define a particular estimation method.  I think it refers to a
general approach and you would need to be more specific about the
criterion.