Dear statalist-participants,
I'd like to estimate my two-level nested logit model using gllamm.
The level-one variables are x1, x2, x3, the level-two-variable is z1 and
the dependent variable on level one is a binary variable y. The
id-variable for level2 is "state".
The model should include a random intercept and a cross-level-effect
between x1 and z1.
Here is the syntax I used:
gen cross=x1*z1
gllamm y x1 x2 x3 z1 cross, i(state) family(binomial) link(logit) adapt
My question is: can I define the cross-level effect in the way I did
(simply by generating a new variable)????? Should I use the option
-eqs- in this case or is it just for modelling random slopes (which I
don't want to model)?
Many thanks for help,
Inna Khousnoullina
*
* 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/
