Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Calculating marginal effects in nonlinear models
From
Boon Han Koh <[email protected]>
To
Statalist Help <[email protected]>
Subject
st: Calculating marginal effects in nonlinear models
Date
Fri, 14 Mar 2014 03:45:00 +0000
Hi Stata users,
I am interested in calculated marginal effects after a probit regression:
Pr(success) = \beta_0 + \beta_1 X1 + \beta_2 X2 + \beta_3 X3
Where X1, X2 and X3 are dummy variables corresponding to treatments in the experiment.
The marginal effect I am interested in is:
\phi( \beta_1 + \beta_2 + \beta_0 ) - \phi( \beta_0 )
In other words, I want to test the joint marginal effect of treatment 1 + 2 relative to the base treatment, evaluated at \beta_3 = 0. For simplicity we can assume that there are no other covariates in the model.
How can I use the -margins- command to achieve this? I know that I can use the "at()" to specify \beta_3 = 0, but I'm not sure how to get Stata to specifically give me the joint marginal effect of \beta_1 + \beta_2. In other words, the difference in predicted probabilities between the baseline and the case where X1 AND X2 = 1.
Regards,
Boon Han
*
* 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/