Re: st: Random Slope, Binary Outcome and AME

[Maarten Buis <maartenlbuis@gmail.com>]

Unfortunately, the answer is none. The marginal effect depends on all
explanatory variables including the community level random effects.
You'll just have to work with odds ratios (and ratios of odds ratios)
as these don't depend on the other variables or random effects.

Hope this isn't too depressing,
Maarten

On Mon, Jan 14, 2013 at 12:37 PM, Thomas Schübel wrote:
> dear all,
>
> i am working with community clustered data on health inequalities. the
> hypothesis is that community (level 2) attributes like local economic
> deprivation are buffering/strengthen the relation between individual income
> and individual health. That all works for continous outcome data (like self
> reported health), but for my binary outome (diagnosis yes-no) i am quite
> confused. i have to report average marginal effects (there might be
> different possibilities, but i definitely have to report them).
>
> Which stata command is working for random intercept / random slope models
> with a binary outcome and can be combined with a command creating average
> marginal effects?
>
> thanks a lot for your help
> thomas
> *
> *   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/



-- 
---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------

*
*   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/