lme4 bugs: ping

Thu, Nov 8, 2007 9:06 PM

I've reported two bugs to this list recently that received no
response:
  lmList does not work for glms (patch included):
     https://stat.ethz.ch/pipermail/r-sig-mixed-models/2007q4/000374.html
  lmer doesn't either:
     https://stat.ethz.ch/pipermail/r-sig-mixed-models/2007q4/000402.html
     https://stat.ethz.ch/pipermail/r-sig-mixed-models/2007q4/000404.html

The first problem is easy enough to fix or work around, just annoying
(but also trivial to fix upstream, since a patch is included!).  The
latter is much more problematic -- lmer is silently giving wildly
invalid results without any indication that they are, in fact,
invalid, and I have no idea how to work around this problem, which
I am encountering while trying to analyze real data that I really
intend to publish once I know how to analyze it.

Doug, I know you're busy, but any chance you could take a look at the
above, esp. the latter?

-- Nathaniel

"The problem...is that sets have a very limited range of
activities -- they can't carry pianos, for example, nor drink
beer."