Hello. Consider a Normal linear mixed effects model where the intercept
has both a fixed effect and random effect. I want to make a confidence interval
for the total intercept=fixed-effect-intercept + random-effect-intercept with
SPlus lme library. Just taking the (fixed-effect +-
1.96*random-effect-standard-deviation) does not take into account the
uncertainty in both estimates.
How can I use the lme library to get a confidence interval that takes
the uncertainties into account? Jose' Pinheiro and Douglas Bates paper "Model
Building for Nonlinear Mixed Effects Models", describes a way to do this based
on a normal approximation and on likelihood profiling, but are they implemented
in the lme library? For the normal approximation, I need the covariance between
the fixed and random effects -- does lme estimate and output this? For
likelihood profiling, is it possible to use profile.nls() with an lme object?
Any help would be appreciated. Thank you for your time,
Hormuzd Katki
Biostatistics Branch, Division of Cancer Epidemiology and Genetics
National Cancer Institute
6120 Executive Blvd. Room 8044 MSC 7244
Rockville, MD 20852-4910
301-594-7818 (voice)
301-402-0081 (fax)
katkih@mail.nih.gov
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|