[R] Re: GLMMGibbs crashes on seeds data

[Jonathan Myles]

Hello Peter,

You haven't included the plate factor in the formula in the glmm
statement, it should be:

glmm(formula=cbind(r,n-r) ~ x1 + x2 + plate, data=seeds, ....)

I thought I had sorted out the problem of models with no random terms,
but clearly I hadn't, so sometime I must try to fix this.  Since
GLMMGibbs uses flat priors for fixed effects, the Credible intervals
for effects in a model with no random effects will be equal to the
frequentist confidence intervals produced by glm()

Best Wishes

Jonathan Myles

On Thu, 30 Aug 2001, Peter Baker wrote:

> Hi all
> 
> I know GLMMGibbs is still in beta but has anyone experienced (and
> solved ;-) this problem?
> 
> I decided to look at the seeds example but I get a core dump on two
> intel linux boxes and also a sun workstation. All are running R1.3.0
> but different hardware/OS's so I think I've done something wrong
> 
> > library(GLMMGibbs)
> > data(seeds)
> > seeds$plate <- as.factor(1:21)
> > seeds$plate <- Ra(data=seeds$plate, shape=0.001, scale=0.001,
> +                   type="identity", contrast="sum", of.interest=F)
> > gseeds.1 <- glmm(formula=cbind(r,n-r) ~ x1 + x2, data=seeds, family=binomial,
> +                store.results=T, icm=500, burnin=1000, keep=20000,
> +                progress.info=500, thin=5)
> Segmentation fault (core dumped)
> 
> > version
>          _
> platform i686-pc-linux-gnu
> arch     i686
> os       linux-gnu
> system   i686, linux-gnu
> status
> major    1
> minor    3.0
> year     2001
> month    06
> day      22
> language R
> 
> $ cat /etc/issue
> Red Hat Linux release 6.2 (Zoot)
> Kernel 2.2.16-3 on an i686
> 
> Suggestions welcome
> 
> Thanks
> Peter
> 
> -- 
> Dr Peter Baker, Statistician (Bioinformatics/Genetics),
> CSIRO Mathematical & Information Sciences, 233 Middle St/
> PO Box 120, CLEVELAND, QLD 4163 Australia.
> Email: Peter.Baker at cmis.csiro.au
> Phone:+61 7 3826 7263 Fax:+61 7 3826 7304
> 
> 
> 
> 


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._