Dear Statalisters,
I have a question concerning how we interpret the coefficients from
-svyintreg-. I have consulted the Statalist archives (I found somebody else
who asked this question, but no replies to it) & faq's, the Stata help files
and the -SVY- manual, and have not found an answer.
My (limited) understanding of the coefficients of -svyintreg- is that they
are not directly interpretable because they are based on a pseudo-likelihood
estimation procedure. Thus, the coefficients predict the effects of changes
in the exogenous variables on the latent variable y*. Consequently,
interpretation is restricted to effect-size and direction.
My question is: can we not calculate the marginal effects at the mean /
media for a specific group of people, thereby giving a more direct
interpretation to the coefficients? If not, why not?
Thanks,
Reza
*
* 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/