Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: IV-GMM or 2SLS

From   nil sen <>
Subject   st: IV-GMM or 2SLS
Date   Sun, 29 Aug 2010 02:03:16 -0500


Please someone help me.

I am estimating a simultaneous equations model. I would like to know
can I use 2SLS or IV-GMM estimation with a binary endogenous
regressor? and with binary dependent variable?


Eq1) Y1=a0+a1*X1+e1,

where X1 is a binary endogenous regressor and Y1 is a continuous
dependent variable.

Which one would be appropriate......IV-GMM: ivreg2 Y1 (X1=IVs for X1),
gmm2s,  or 2SLS:  ivreg2 Y1 (X1=IVs for X1)?

Eq2) X1=b0+b1*Y1+e2.

where X1 is a binary dependent variable and Y1 is continuous.

Which one would be appropriate......IV-GMM: ivreg2 X1 (Y1=IVs for Y1),
gmm2s,  or 2SLS:  ivreg2 X1 (Y1=IVs for Y1)?

All the instruments for X1 and Y1 are dummy variables.

If somebody knows the answer, please help.

An early reply is eagerly awaited.

Thank you very much.


*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index