st: non-linear regression with mixed cross-sectional and longitudinal data

[Kirstie Whitaker <kirstie.whitaker@berkeley.edu>]

Hi Stata-list,

I'm a long time listener, first time caller.  Thank you for all the
help you've provided over the years.

I'm calculating regression models of the form

y = a + b * exp( -x / c)

(where a, b and c are the parameters I'm looking to estimate)

with the command:

nl ( WMI = {alpha} + {beta}*exp(-Age/{time=1}) )

(If you're interested - this is looking at the developmental
trajectory of white matter in the adolescent brain)

My data set has some participants who have been in for two scans,
about 18 months apart.  I'd like to be able to include all the data in
my model, using for example xtmixed with a random effect of subjectID
or regress with the cluster(subID) option however I'm not sure how to
fit the hypothesized model (see above) into either of these
frameworks, or if I can use nl with a random effect.

Does anyone have an idea?  I wonder if there is a way to model this in
a linear way that I'm just not seeing?

Thank you for all the help!
Kx

-- 
Thank you!  Together we've raised $13,000 over the last four
AIDS/Lifecycle rides. I couldn't make it along those 545 miles from SF
to LA without your support.
-----------------------
Kirstie Whitaker
Doctoral Candidate
Bunge Laboratory: Building Blocks of Cognition
Helen Wills Neuroscience Institute
University of California at Berkeley
134 Barker Hall, MC 3190
Berkeley, CA, 94707
tel: 510 684 2456
web: bungelab.berkeley.edu

*
*   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/