Re: Re: st: Interpretation of Oaxaca decomposition results after re-transformation of log scale

[Vaidyanathan Ganapathy <vaidyang@usc.edu>]

Hi Nichols,

Thanks for your response about the Oaxaca interpretation. I am
familiar with the generalized linear modeling techniques. But my
understanding is that GLM fits population average models or at best
random effects/mixed models. My ultimate goal is to fit fixed effects
models to account for unobserved individual specific characteristics.
To be specific, I want to run separate fixed effects models among
control and premature infant cohorts and decompose the differences.

To answer your question about still births, No, my sample includes
only children surviving beyond 6 months of age.

Thanks !
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/