Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Overidentification test after cmp command


From   v nguyen <katiana00@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Overidentification test after cmp command
Date   Tue, 22 Sep 2009 10:27:42 -0700 (PDT)

Hello,

I used a cmp command (written by David Roodman) to estimate an IV probit where the endogenous variable is binary. My model is overidentified, i.e. I have one suspected endogenous variable and 2 excluded instruments. I now would like to carry out an overidentification restriction test to see if one of my IVs is indeed a "good" IV. In short, I would like to use an "overid" command (written by Kit Baum & coauthors), but this is not available after issuing cmp.

To be concrete, this is my code now and I want to test that UNEMR_NSA is properly excluded from the D equation

svy: cmp (D: MRJMON = UNEMPL MRJPRICE HERPRICE METHPRICE IRSEX MAR IRAGE IRAGE2 IREDUC HS NCHILD MSA CITY RURORURB NONWAGE) (E: UNEMPL=N_W UNEMR_NSA MRJPRICE HERPRICE METHPRICE IRSEX MAR IRAGE IRAGE2 IREDUC HS NCHILD MSA CITY RURORURB NONWAGE) , ind($cmp_probit $cmp_probit)

Thank you for any help.

Vuong


      
*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index