[R-sig-ME] mcmcsamp can't handle random effects syntax

[Nicholas Lewin-Koh]

Duhh, that should have occurred to me, if I'd just looked at the model
ouput. Thanks

Nicholas

-----Original Message-----
From: dmbates at gmail.com [mailto:dmbates at gmail.com] On Behalf Of Douglas
Bates
Sent: Wednesday, September 17, 2008 9:11 AM
To: Nicholas Lewin-Koh
Cc: r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] mcmcsamp can't handle random effects syntax

On Wed, Sep 17, 2008 at 11:05 AM, Nicholas Lewin-Koh
<lewin-koh.nicholas at gene.com> wrote:

> Hi,
> This may be well known, but it was mucking me up.

>> fm1 <- lmer(Reaction ~ Days + (1|Subject) + (0+Days|Subject), sleepstudy)
>> ttt<-mcmcsamp(fm1,n=1000)
>> fm1 <- lmer(Reaction ~ Days + (1+Days|Subject), sleepstudy)
>> ttt<-mcmcsamp(fm1,n=1000)
> Error in .local(object, n, verbose, ...) :
>  Code for non-trivial theta_T not yet written
>> fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)
>> ttt<-mcmcsamp(fm1,n=1000)
> Error in .local(object, n, verbose, ...) :
>  Code for non-trivial theta_T not yet written

> These are all the same model, right?

Not really.  The first model has uncorrelated random effects.  The
other two have correlated random effects.  The rather cryptic and
uninformative error message is a reminder to me that I need to write
code for the MCMC update of a covariance parameter.