State of lme4

Out of curiousity, why not just use the asymptotic standards errors of
the fixed effects to get Cis rather than via simulation?
-----Original Message-----
From: r-sig-mixed-models-bounces at r-project.org 
[mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf 
Of Jonathan Baron
Sent: Thursday, July 10, 2008 8:05 AM
To: Michael Kubovy
Cc: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] State of lme4

On 07/10/08 07:51, Michael Kubovy wrote:
Dear Friends,

I have become confused as to which set-up of lme4, arm, 
gmodels, etc 
will produce CIs on the fixed effects by simulation.
In the latest version of lme4, mcmcsamp and HPDinterval work. 
 And languageR has a new version as of yesterday, which seems 
to deal correctly with the current version of lme4.  
Specifically, pvals.fnc() works.  I don't know about arm and 
gmodels.  I haven't seen new versions, so I suspect they will 
not deal with the new format of lmer.

Jon
--
Jonathan Baron, Professor of Psychology, University of 
Pennsylvania Home page: http://www.sas.upenn.edu/~baron

_______________________________________________
R-sig-mixed-models at r-project.org mailing list 
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

State of lme4

Thread (5 messages)