[R] problem with lme in nlme package

[Douglas Bates]

John,

I'm travelling today and won't be able to check this discrepancy.  I
will try to look at it tomorrow.

A couple of things to consider:

 - the underlying optimization software is different in S-PLUS and R.
   It may be that the REML estimates are poorly defined and
   convergence is being declared in different places by different software.
   Have you checked the intervals on those variance components using
   intervals(bryk.lme.1)

 - I believe the data are already in the nlme package as the
   MathAchieve and MathAchSchool data sets.

 - In Bryk's analysis the meanses is calculated incorrectly.

 - Some analysis of these data is given in the slides at
   http://www.stat.wisc.edu/~st850-1/MEcourse6_4up.pdf

 - It is much easier (and faster) to calculate the meanses gsummary.
   I think it would be gsummary(MathAchieve, which = "ses")


-- 
Douglas Bates                            bates at stat.wisc.edu
Statistics Department                    608/262-2598
University of Wisconsin - Madison        http://www.stat.wisc.edu/~bates/
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._