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]

Re: st: estimating treatment effect of a binary endogenous regressor on binary outcome


From   Erkan Duman <[email protected]>
To   [email protected]
Subject   Re: st: estimating treatment effect of a binary endogenous regressor on binary outcome
Date   Fri, 15 Nov 2013 17:57:05 +0200

Austin- i may have explained my problem in a wrong fashion. The
treated households are migrant househols- househols where at least one
of the parents is absent due to migration. I control being a migrant
household by using a dummy which takes value 1 if at least one of the
members of the household receives remittances in the last year.
I use regional migration rates as an instrument which captures the
strength of the migration networks. In a high migration rate region,
the cost of migrating is lower and it increases the probability of
receiving remittances for households in that region. The instrument is
very strong with an F Statistic of 31.
Jorge- the evidence on estimated effects of migration on student
attendance is mixed but generally it is negative or a very low
positive effect. Observing 80 percent increase in school attendance
for migrant households is therefore seem too much to me. And i should
not get a coefficient over 1 since it i a probability model but
getting 10 will signal a problem in estimation method or my
specification.
Nowadays i am thinking of my instrument being invalid. Chiburis
suggested me that this may be the cause of getting a coefficient over
1. I am considering controlling more variables that may be correlated
with my instrument and have a direct effect on the outcome. I hope
this will solve my problem.
Thank you for your answers.

On Wed, Nov 13, 2013 at 5:58 PM, Austin Nichols <[email protected]> wrote:
> Erkan Duman <[email protected]>:
>
> See also
> http://www.stata-journal.com/article.html?article=st0144
> http://www.stata.com/meeting/new-orleans13/abstracts/materials/nola13-barker.pdf
>
> But what kind of endogeneity are you thinking you will correct?
> With what instruments?
> The story you tell for the LATE you are estimating and why it is
> identified is at least as important as any statistic your program
> spits out.
> Tell the story as a thought experiment... e.g.
> you imagine dropping a migrant Mexican parent on a randomly selected
> US resident child, or randomly assigning a Mexican family to move to
> the US instead of remaining in Mexico. (Replace Mexico with Turkey and
> the US with Germany, or what have you.)
>
>
> On Wed, Nov 13, 2013 at 4:22 AM, Erkan Duman <[email protected]> wrote:
>> Hello.
>> I am working on the impacts of having a migrant at home on the school
>> attendance of children. My regressor is an endogenous binary variable
>> and the outcome is also binary. I have searched a lot to find an
>> estimator which will consistently estimate the treatment effect.
>> Chiburis et al. (2012) paper recommends bivariate probit or IV2SLS.
>> However, bivariate probit requires strong assumptions (bivariate
>> normal errors) and my specification does not satisfy this assumption
>> which results in severly biased estimates. IV2SLS gives a treatment
>> effect over 1 (actually around 10). In my study, the treatment
>> receivers only constitute around 2 percent of the sample. I believe
>> this low treatment probability causes problems in estimating the
>> treatment effects because when I reduce the size of the control group
>> and come up with a treatment group of 16% of the sample, the treatment
>> estimate fits in (0,1) range ; still it is too high around 0.80. I
>> have encountered a similar thing to my problem as "rare events" in the
>> literature, but doesn't seem to solve my problem or maybe I am wrong.
>> Also in http://www.stata.com/meeting/chicago11/materials/chi11_nichols.pdf
>> some semi-parametric estimators are suggested by  Austin Nichols which
>> do not require bivariate normal errors but weaker assumptions.
>> However, I am not familiar with semi-parametric or nonparametric
>> estimators. Can anyone help me with an appropriate semiparametric
>> estimator and its stata command?
>> I believe there is a solution to this estimation problem and I hope
>> someone will help me.
>> Thanks.
>> Best regards.
>>
>> --
>> Erkan Duman
>> Graduate student - PhD
>> Faculty of Art and Social Sciences
>> Sabancı University
>>
>> *
>> *   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/
>
> *
> *   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/



-- 
Erkan Duman
Graduate student - PhD
Faculty of Art and Social Sciences
Sabancı University

*
*   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