[Limdep Nlogit List] Odds ratios

Steven Yen syen at utk.edu
Thu Apr 9 07:25:09 EST 2009


I estimated a (bivariate) generalized logistic model. I am too 
obsessed with marginal effects on marginal and conditional 
probabilities, but a nutrition journal prefers odds ratios.

Specifically, I reported the marginal effects of each variable on the 
following marginal and conditional probabilities:

   Pr(d1=1)
   Pr(d2=1|d1=1)

Personally, I might differentiate (and differencing, for discrete 
variables) what I know to be the odds ratios

   Pr(d1=1) / [1-Pr(d1=1)]
   Pr(d2=1|d1=1) / [1 - Pr(d2=1|d1=1)]

BUT, any reference to suggest?



Steven T. Yen
http://web.utk.edu/~syen/
Check out our postdoc opening at:
http://web.utk.edu/~syen/postdoc.html



More information about the Limdep mailing list