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: test interaction effects after heckprob
From
Marco Sangiorgi <[email protected]>
To
<[email protected]>
Subject
st: test interaction effects after heckprob
Date
Wed, 22 Jun 2011 10:38:43 +0200
Hi All,
I'm interested in testing significance of interaction terms in a heckprob model of the type:
y_p i.x1 c.x2 i.x1#c.x2 ..., (select y_s i.x1 c.x2 i.x1#c.x2 ...) robust
with x1=dummy variable.
All the contribution I've red explain how to do it in a simple probit model (e.g. -inteff-) but not in a two stage model like this. A simple t-test on the coefficent of the intaraction is not informative (Ai and Norton, 2003).
Since I'm interested in analysing the joint probability p(11) I used -margins- to predict such probability for (x1= 0) and for (x1=1) and so I tested for their difference, but I don't think it's a good way to obtain reliable statistics.
Any suggestion?
Thanks a lot.
Marco Sangiorgi
*
* 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/