Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Predicted probabilities from gllapred using mlogit


From   Joanne Flavel <flav0031@flinders.edu.au>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: Predicted probabilities from gllapred using mlogit
Date   Thu, 16 Aug 2012 19:23:47 -0700

Dear Statalisters
I have estimated a 2 level multinomial logit for a model with a categorical dependent variable with 3 outcomes using gllamm . I am now trying to estimate average predicted probabilities but am having some trouble so was hoping for some advice

My model is given by:

gllamm depvar varlist, i(id) nrf(1) eqs(cons) base(2) link(mlogit) family(binom) robust trace

I then used gllapred

gllapred prob1, mu marginal outcome(1)
gllapred prob2, mu marginal outcome(2)

Outcome 3 is my base outcome

I understand that for a logit, the average predicted probability can be obtained by the following:

gen phat=exp(prob1)/(1+exp(prob1)

I am stuck on how to calculate this for the multinomial logit. Can anybody please advise on the correct way to do this? Any help will be much appreciated

Kind regards
Joanne


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


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index