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 difference MEM AME
From
Luca Fumarco <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: test difference MEM AME
Date
Sat, 10 Nov 2012 18:04:02 +0100
Dear Statalisters,
I am using a heteroskedastik probit model. I want to gain the marginal effect of a dummy variable.
Because of the empirical approach that I decided to adopt in my paper, I want to compute such a marginal effect with the derivative instead of the difference.
Since I use the hetprob model, I cannot use Bartus' command "margeff"; instead, I think to use margins and margins at mean, so that I get both the MEM and the AME wrt my dummy variable of interest.
(Tamás Bartus, "Estimation of marginal effects using margeff" Stata Journal (2005), Vol. 5, Number 3, pp. 309–329)
Once I have gotten these two estimates I wwnt to test whether they are statistically different.
So, how can I test whether the MEM and the AME obtained with the margins are statistically different?
Any suggestion?
Thank you in advance,
have a nice w-e
Luca
*
* 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/