I don't see enough information for me to diagnose the problem. If
it were my problem, I'd print out some assessment of the results of
every step you've described below. This should expose the step in which
things disappear.
Hope this helps.
Spencer Graves
p.s.
OK and thanks!! I will try to apply the approach you suggest to my data.
I found a possibly interesting reference on this topic (if my fast reading
is error-free):
E. Demidenko, 2004, Mixed Models: Theory and Applications (John Wiley
Sons), Chapter 9 (Diagnostics and influence analysis)
Spencer,
A number of times I've been wrecking my head trying to solve a
problem, and when I thought I'd tried everything possible (all the
docs that I can think of, web searches etc...) I start re-reading the
posting guide (I've no idea how many times I've read it now)... and 9
times out of 10,
101 - 103 of 103 matches
Mail list logo