[R-sig-ME] lme4, failure to converge with a range of optimisers, trust the fitted model anyway?

[Hans Ekbrand]

On Sun, Apr 05, 2015 at 01:56:03PM +0200, Hans Ekbrand wrote:
> > > > While I'm thinking of it, should clusters and country random effects have
> > > > been crossed. Generally the sampling is setup so that clusters are nested
> > > > within countries which requires a different syntax.
> > >
> > > I'm sorry but I haven't been clear on this, but the clusters are
> > > nested within countries, so there are no crossed random effects to be
> > > found.
> > >
> > >
> > The  random effect then needs to be included as (1|country/clusterID)
> 
> Well, no, cluster is implictly nested in country, just as sample is
> implicitly nested in batch in the Pastes data (see p 39-40 in
> http://lme4.r-forge.r-project.org/book/Ch2.pdf)

Sorry if this was unclear, but the ClusterID variable was created in a
similar way that Bates creates the sample variable on page 40 in that
text. [ English is not my native language ]