st: unbalanced by design

[Clinton Thompson <clintonjthompson@gmail.com>]

Hello All,
I have longitudinal data in the sense that my outcome variable was
assessed at three time points (baseline, month 1, month 2) and I'm
interested in analyzing how this variable is affected by a number of
other covariates, some of which were assessed (by design) only at
baseline, at baseline and month 2, and the remaining at all three time
points.  I've scoured the StataList archives, skimmed through my
textbooks, and otherwise racked my brain for a way to analyze this
data in such a way as to incorporate all the data in the various
models, even when the covariate(s) was/were assessed at just baseline.
 I've considered a couple of different ways to analyze the data (e.g.
-xtreg-, -xtgee-) and although they perform as expected -- those data
points that were assessed at month 1 and month 2 for the outcome but
weren't assessed for the independent variable are omitted from the
model -- I'm wondering if there is another approach that incorporates
more (all?) of the data.  I realize this isn't a Stata question per
se, but given the richness of Stata's longitudinal/panel data modeling
capabilities, I'm hopeful that a Statalist member may be able to point
me in the right direction.

I'm using Stata/SE 11.2.

Many Thanks!
Clint Thompson
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/