Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Finding the interaction effect after ivprobit


From   "Kelvin Foo" <kelvinski@gmail.com>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: Finding the interaction effect after ivprobit
Date   Sat, 2 Sep 2006 22:01:46 +0100

Dear Statalisters,

I have an -ivprobit- estimation which contains an interaction term:

ivprobit dvar (var1=var2) var3 var4 var5

where var5 is the interaction term generated from var3*var4

As far as I know, there is no single command to find the interaction
effect and its standard errors for the above in Stata 9.

Doing so by hand, is the interaction effect in this case the same way
as in a simple probit model? That is, differentiating

Normal(_b[_cons]+_b[var1]*var1+_b[var3]*var3+_b[var4]*var4+_b[var5]*var3*var4])
with respect to var3 and then var4. Is this correct?

For the standard errors, will it be correct to run

predictnl interact=<interaction effect>, se(se)

with the se variable giving the standard errors?

I am unsure if my procedure is right, especially since the -ivprobit-
MLE process involves more than one equation.

Would greatly appreciate any comments as well as further suggestions.
Thanks in advance!

Kelvin
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index