Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: GLLAMM repeated measures interaction problem


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   stata list <statalist@hsphsun2.harvard.edu>
Subject   st: RE: GLLAMM repeated measures interaction problem
Date   Mon, 20 Apr 2009 13:08:26 +0000 (GMT)

--- Marnie wrote me privately:
> I am running a 2-level random intercept logistic model using gllamm
> (reshaped data). I am investigating the effect of the interaction
> of dose*time on the nominal dependent variable controlling for
> school level clustering. Dose is continuous (0-100%) and time is
> categorical (4 time points). I am receiving the error message
> "variables have been dropped, can't continue. I did try to create
> my own dummy variables however it did not create the variables for
> each time point.
> 
> My syntax is below. The other variables in the model are
> confounders.
> xi:gllamm q9nom i.chgender i.bord_3 i.empl3_ i.q4bbin i.educ4_
> i.time*Dose if(walked==1), i(studid schid) base(1) link(mlogit)
> family(binom) nip(7) adapt

First, the best way of getting advice is to contact the statalist
directly, as is discussed in the Statalist FAQ:
http://www.stata.com/support/faqs/res/statalist.html#private

Second, I don't know the answer (thus proofing my first point).
If I would run into a problem like that I would probably first
try to run the model with -mlogit- rather than -gllamm-. 
Simpler models are usually easier to debug than a full blown 
-gllamm- model. 

-- Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index