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: ivprobit and hectroskedasticity


From   Abekah Nkrumah <ankrumah@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: ivprobit and hectroskedasticity
Date   Fri, 4 May 2012 15:28:48 +0100

Dear Stata List,

I am running an ivprobit. However, my understanding is that  ivprobit
with an mle or twostep option is meant for homoskedastic errors.My
data is collected via stratified sampling and so there will be
intra-cluster correlation and thereby violating the homoskedasticity
assumption.

My thinking is to adjust the standard errors using the cluster robust
estimator in the case mle option and a bootstrap in the case of
twostep

However I am not too sure whether the associated IV diagnosis (eg.
overidentification test, Endogeneity test, weak instrument test etc)
from the model will be consistent for a hectroskedastic case?

I understand that in the case of linear models a solution will be to
use a GMM estimator which is consistent in the presence of
hectroskedasticity. Does this apply in the case of a probit model as
well?

I ask this question knowing that there insn't a straight forward
command for GMM probit, but I know that it can be sorted out manually
from stata.

Thanks very much

Regards
*
*   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/


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