--------------------------------------------------------------------------------
My immediate reaction would be to say that it is not possible within
-gllamm-. The two-way effects (like country and industry) are quite
tricky for -gllamm- per se, and I would suspect there is no way at all
to specify the product of two random effects.
So may be you would have to think differently about what exactly you
want to model. If all you want are random effects for each combination
of the county and industry (as it seems to be the case), then
egen id = group(cc industry)
gllamm y , i(id) link(ologit)
You won't be regressing anything on the country ID in your list,
anyway, would you?
Stas
On Sat, 02 Oct 2004 05:23:21 +0000, Dana Shills <[email protected]>
wrote:
>Hi
>
>Im new to GLLAMM analysis and need some help figuring out if I am doing the
>right thing. I am trying to fit a random effects model to the following
>ordinal response data:
>Country(1-80) Industry(1-3) Firm Y(1-5)
>1 1 1 3
>1 1 2 2
>1 1 3 1
>1 2 4 2
>1 2 5 3
>2 1 6 5
>2 1 7 4
>2 2 8 2
>2 3 9 4
>.....
>For the following model,
>Yij=alpha+COUNTRYi+INDUSTRYj+COUNTRYi*INDUSTRYj+Eij
>assuming ONLY random effects due to country, industry and the
>country-industry interactions, I want to obtain variance components
>estimates due to country, industry, the interactions and the residual
>variance. How should I go about doing this? Would it be:
>
>gen C_I=cc*industry
>eq int : const
>eq C : cc
>eq I : I
>eq C_I : C_I
>
>gllamm y C I C_I, i(country) eqs(int C I C_I) link(ologit) adapt trace
>
>I would really appreciate some help on this.
>
>Thanks
>
>Dana
>
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
*
* 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/
