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

[Douglas Bates]

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.