[R-sig-ME] R-sig-mixed-models Digest, Vol 15, Issue 23
Iasonas Lamprianou
lamprianou at yahoo.com
Fri Mar 28 13:05:01 CET 2008
Prof. Bates,
we all thank you for investing so much time to write the book (Multilevel models in R). A polite request: please, please include as many practical and detailed examples as possible, Not just one example per chapter. And what we need, is detailed guides on how to interpret the results. And how to defend our models: i.e. to prove that they have satisfactory fit and that we have not violated their assumptions.
Thanks
Dr. Iasonas Lamprianou
Department of Education
The University of Manchester
Oxford Road, Manchester M13 9PL, UK
Tel. 0044 161 275 3485
iasonas.lamprianou at manchester.ac.uk
----- Original Message ----
From: "r-sig-mixed-models-request at r-project.org" <r-sig-mixed-models-request at r-project.org>
To: r-sig-mixed-models at r-project.org
Sent: Friday, 28 March, 2008 1:00:02 PM
Subject: R-sig-mixed-models Digest, Vol 15, Issue 23
Send R-sig-mixed-models mailing list submissions to
r-sig-mixed-models at r-project.org
To subscribe or unsubscribe via the World Wide Web, visit
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
or, via email, send a message with subject or body 'help' to
r-sig-mixed-models-request at r-project.org
You can reach the person managing the list at
r-sig-mixed-models-owner at r-project.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of R-sig-mixed-models digest..."
Today's Topics:
1. Re: New alpha version of lme4 code on R-forge (Douglas Bates)
2. use of pedigree structures (ian white)
3. Re: use of pedigree structures (Douglas Bates)
4. Re: how to locate standard errors in lmer output (Douglas Bates)
----------------------------------------------------------------------
Message: 1
Date: Thu, 27 Mar 2008 08:17:59 -0500
From: "Douglas Bates" <bates at stat.wisc.edu>
Subject: Re: [R-sig-ME] New alpha version of lme4 code on R-forge
To: "Dan Wright" <danw at sussex.ac.uk>
Cc: R Mixed Models <r-sig-mixed-models at r-project.org>
Message-ID:
<40e66e0b0803270617v4bd360a6j9987e01f005a9a27 at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
On Thu, Mar 27, 2008 at 4:30 AM, Dan Wright <danw at sussex.ac.uk> wrote:
> Dear Professor Bates,
> Thank you for this! I have some questions that I assume others on the list
> are interested in. You mentioned in one message that you are working on a
> book. What is it on? When out? Is it appropriate to cite when using lme4?
Thanks for the inquiry. The working title of the book is "Multilevel
modeling in R" although that may change before publication. It is on
the practice of using the lme4 package to fit and analyze
mixed-effects models, including linear mixed models, generalized
linear mixed models, nonlinear mixed models and generalized nonlinear
mixed models. I emphasize graphical displays of the data and aspects
of the fitted models and inferences based on MCMC samples from the
posterior distribution of the model parameters. The statistical
theory and computational methods for fitting and analyzing the models
are also described in detail.
I have given up on issuing predictions of when the book will be
available. "Not soon" is about the best I can say. There is a lot of
writing still to be done. Thanks for the suggestion of citing the
book when using lme4 but that would be premature.
------------------------------
Message: 2
Date: Thu, 27 Mar 2008 12:40:48 +0000
From: ian white <i.m.s.white at ed.ac.uk>
Subject: [R-sig-ME] use of pedigree structures
To: r-sig-mixed-models at r-project.org
Message-ID: <47EB95D0.3070602 at ed.ac.uk>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
The lmer function includes random terms as
(expression for a model matrix|grouping factor)
and I wonder how general the expression can be. For example, a pedigree
object can be converted to an A matrix (in the language of animal
breeding), but can the A matrix be incorporated into an lmer analysis?
I thought that
(chol(pedigree)|ones)
might be OK, where 'ones' is a constant vector of value 1. With the only
example I have tried, the syntax is accepted but the analysis ends with
a forced exit from R (core dump).
--
I White
University of Edinburgh
Ashworth Laboratories, West Mains Road
Edinburgh EH9 3JT (Tel 0131 650 5490)
------------------------------
Message: 3
Date: Thu, 27 Mar 2008 11:11:45 -0500
From: "Douglas Bates" <bates at stat.wisc.edu>
Subject: Re: [R-sig-ME] use of pedigree structures
To: "ian white" <i.m.s.white at ed.ac.uk>
Cc: r-sig-mixed-models at r-project.org, Ana Vazquez
<anainesvs at gmail.com>
Message-ID:
<40e66e0b0803270911o114c58dcv72b5b1f14f2332bf at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
The pedigree class has been moved to the pedigreemm package, which is
also on R-forge
http://r-forge.r-project.org/projects/pedigreemm
In that package Ana Vazquez and I are developing methods to
incorporate pedigrees using animal models or sire models. That
package is still under development. If you are particularly anxious
to use those methods you may want to contact Ana directly.
On Thu, Mar 27, 2008 at 7:40 AM, ian white <i.m.s.white at ed.ac.uk> wrote:
> The lmer function includes random terms as
>
> (expression for a model matrix|grouping factor)
>
> and I wonder how general the expression can be. For example, a pedigree
> object can be converted to an A matrix (in the language of animal
> breeding), but can the A matrix be incorporated into an lmer analysis?
>
> I thought that
>
> (chol(pedigree)|ones)
>
> might be OK, where 'ones' is a constant vector of value 1. With the only
> example I have tried, the syntax is accepted but the analysis ends with
> a forced exit from R (core dump).
>
>
> --
> I White
> University of Edinburgh
> Ashworth Laboratories, West Mains Road
> Edinburgh EH9 3JT (Tel 0131 650 5490)
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
------------------------------
Message: 4
Date: Thu, 27 Mar 2008 11:18:36 -0500
From: "Douglas Bates" <bates at stat.wisc.edu>
Subject: Re: [R-sig-ME] how to locate standard errors in lmer output
To: "F.-Xavier ALARIO" <Francois-Xavier.Alario at univ-provence.fr>
Cc: R Mixed Models <r-sig-mixed-models at r-project.org>
Message-ID:
<40e66e0b0803270918n2d422fe3ta6cd1f7e322911af at mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Thu, Mar 27, 2008 at 5:06 AM, F.-Xavier ALARIO
<Francois-Xavier.Alario at univ-provence.fr> wrote:
> I have been looking in various documentations (e.g. in
> http://rweb.stat.umn.edu/R/library/lme4/doc/) for a way to access
> Standard Errors of fixed effects in the output of the 'lmer' function.
> slotNames() didn't help.
> Is there a way to do it similar to fixef(), ranef() or with a slot name?
The preferred way would be
stderr <- sqrt(diag(vcov(fm)))
for a single evaluation. (If you were putting such a calculation
within a large simulation you may want to work harder to tune it up
but this is fine for a single evaluation.)
> I have the same question for t-values.
fixef(fm)/stderr
The obvious next question is about the p-values but that's a sore
point so don't ask. :-)
------------------------------
_______________________________________________
R-sig-mixed-models mailing list
R-sig-mixed-models at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
End of R-sig-mixed-models Digest, Vol 15, Issue 23
**************************************************
_______________________________________
cs.yahoo.com/nowyoucan.html
More information about the R-sig-mixed-models
mailing list