Dear all,Back to the Top
When we can use SAS PROC GLM and SAS PROC MIXED
When the design of the study is 2*2 3*3.
When the design of the study is replicated.
It will be nice any one could suggest me that am I correct or not.
Back to the Top
Replication has nothing to do with whether to use GLM or Mixed. If the
data are complete and no missing data then either can be used, although
GLM will be faster. GLM uses an exact solution to the linear system of
equations, whereas Mixed uses an iterative process. If the data are
unbalanced and you have missing data then Mixed is the way to go because
of the way missing data are handled with GLM. GLM, I believe, deletes
the entire subject from an analysis if that person has any missing data,
but Mixed does not. You are also more restricted with GLM in that
compound symmetry is assumed. With Mixed this assumption is relaxed and
more generalized covariance matrices can be used.
Also, the BE statistical guidance used Mixed as an example, not GLM.
Hope this helps,
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (firstname.lastname@example.org)