Another case of -1.0 correlation of random effects
Andrew Dolman wrote:
Is it possible to post your data, or a dummy data set, so that we can
definitely rule out a structural problem? Do you get the same problem
with a dummy dataset?
The function dump("myData", file=stdout()) might be useful.
Andy.
andydolman at gmail.com <mailto:andydolman at gmail.com>
You should be able to get it with
load(url("http://www.kevinthorpe.ca/~kevin/gluc.RData"))
The profile function did work with the Diestuff example in chapter 1 of
the lme4 book, so I know that profile() is working.
Kevin E. Thorpe Biostatistician/Trialist, Knowledge Translation Program Assistant Professor, Dalla Lana School of Public Health University of Toronto email: kevin.thorpe at utoronto.ca Tel: 416.864.5776 Fax: 416.864.3016