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: Abadie's Causal IV method in Stata


From   Karan Nagpal <karan.nagpal1@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Abadie's Causal IV method in Stata
Date   Wed, 18 Apr 2012 11:39:18 +0100

Hello all,

I'm trying to estimate causal impacts in a binary dependent variable,
binary endogenous variable situation. One method recommended by
Angrist is Abadie's Causal IV method, that uses some kind of a
weighting scheme to estimate causal impacts using least squares.

If I have understood correctly, what it implies is estimation through
minimising least squares in which each observation's least square is
"weighted" (by kappa, a factor that allows us to identify impact among
the compliers). However, some of the individual weights can be
negative.

How to implement this in stata? I thought of using the weighted least
squares command, wls0, but that returns an error: "time series
estimators not allowed." I guess this happens because some of the
kappas are negative. Another idea is to estimate using probit, using
the weight command. My probit regression didn't achieve convergence
for more than 4000 iterations so I gave up.

Is there any way that I can estimate "weighted" least squares where
weights can be negative?
*
*   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