Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: Testing equality across two specifications of ivreg2


From   Jessica Ellen Leight <jeleight@mit.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Testing equality across two specifications of ivreg2
Date   Wed, 05 May 2010 12:34:23 -0400

Dear all,

I am estimating two different instrumental variables specifications using ivreg2
on the same sample.  I have observations on two individuals per household (call
them x1 and x2), each instrumented with an individual-level shock, and I
estimate first the household fixed effects specification using the difference
across individuals.

ivreg2 ydif (xdif=shockdif), cl(village year)

Subsequently, I estimate an individual's outcome as a function of both x1 and
x2, appropriately instrumented, without household fixed effects.

ivreg2 y1 (x1 x2=shock1 shock2), cl(county year)

I now wish to test whether the coefficient on x2 in the second specification is
equal to the coefficient on xdif in the first specification, or (a second test)
whether _b[xdif]=_b[x2]-_b[x1].  Suest does not work smoothly with ivreg2 as
others have noted, and previously suggested solutions on this list all seem to
apply to cases in which the same specification is being estimated on two
different subsamples, rather than two slightly different specifications with
the same sample.

Any suggestions are appreciated.
Best,
JL
*
*   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   |   Site index