st: RE: Endogenous truncated variable

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Jose,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> [iso-8859-1] Jos� Ignacio Ant�n
> Sent: Thursday, August 28, 2008 11:13 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Endogenous truncated variable
> 
> Dear Stata users,
> 
> I want to estimate a model where I have a truncated 
> endogenous variable (remittances, which is 0 for 90% of 
> households). I want to instrument this variable (possible two 
> way causality and other problems) and I would like to know if 
> 2SLS can be appropriate. The dependent variable is continuous.
> 
> I think some variations of this question have been posted 
> before, but there were very different opinions.

Which different options do you mean?  I've answered this question on the list on a number of occasions, and the answer I give doesn't change. :)  E.g.:

http://www.stata.com/statalist/archive/2004-07/msg00710.html
http://www.stata.com/statalist/archive/2007-04/msg00945.html

And agrees with those of others, e.g.:

http://www.stata.com/statalist/archive/2004-09/msg00337.html

For a RHS endogenous regressor, IV is fine whether or not the variable is truncated, binary, whatever.  You can get efficiency gains over IV using system estimation methods that take account of the form of the regressor, but IV is still consistent.  More precisely, if it's well specified when it's exogenous, it's also well specified when it's endogenous and estimated by IV with the usual conditions on the instrument(s).

Cheers,
Mark

> I will 
> appreciate an answer since it is a question of a dissertation 
> I must submit in the next days.
> 
> Thank you very much,
> 
> Jos� Ignacio Ant�n
>  
> 
> __________ Informaci�n de ESET Smart Security, versi�n de la 
> base de firmas de virus 3341 (20080808) __________
> 
> ESET Smart Security ha comprobado este mensaje.
> http://www.eset.com
>  
> 
> 
> 
> 
> *
> *   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/
> 


-- 
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   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/