Bookmark and Share

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: endogenous regressor in bioprobit and cmp


From   "Elena Pirani" <[email protected]>
To   [email protected]
Subject   st: endogenous regressor in bioprobit and cmp
Date   Thu, 29 Aug 2013 14:55:00 +0200

Dear Statalist users,
I am trying to run a bivariate ordinal probit model, i.e.:
y_1=x1 x2 x3 x4
y_2=x1 x2 x3 y_1
that is a bivariate probit with an endogenous regressor (y_1) and where x4 is my
instrumental variable for y_1.
Following bioprobit paper and routine of Sajaia (2008) it seems that the correct
sintax is:

bioprobit (y_1 = x1 x2 x3 x4 )  (y_2 = x1 x2 x3) , end
that enables to obtain the parameter gamma.

However if I add the endogenous variable y_1 in the second equation, i.e.
bioprobit (y_1 = x1 x2 x3 x4 )  (y_2 = x1 x2 x3 y_1)
without the option "end"  I get estimates that are different from the first model
specification.
I obtain an estimation for the parameter y_1 in the second equation that is different
from gamma in the previous formulation (and in this case I don't have gamma!)

Could someone help me? Which is the difference? Which is the more correct approach?
Moreover, does someone knows why the option "end" described in the paper of Sajaia
is not present in the current version of online help of Stata?

Please note that when using cmp sintax
cmp (y_1 = x1 x2 x3 x4 )  (y_2 = x1 x2 x3 y_1) ...
I obtain estimation coefficient similar to the bioprobit without "end" option.

I would greatly appreciate any assistance which you might be able to provide.
Thanks
Elena
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index