Shige Song wrote:
Can xtmixed be used to estimate multilevel growth model with AR(1)
disturbance term, like what one can do with the REPEAT option in Proc
Mixed in SAS? A quick glance on the manualI did not find any
discussion on this topic. Thanks!
----------------------------------------------------------------------------
It doesn't seem like any such patterned covariance structures are built in.
Haven't thought about it much, but is there any chance that you could set up
a block-diagonal structure with or without the use of -eqn- or -constraint-
to get what you want?
As for first-up enhancements to -xtmixed-, I vote for denominator degrees of
freedom calculations (esp., Kenward-Roger) over further patterned covariance
structures, implementing a -robust-/cluster()- option, or building upon it
with GLM-like transformations � la PROC GLIMMIX.
Joseph Coveney
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
