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 on April 23, and its replacement, statalist.org is already up and running.


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

st: RE: 2SLS with selected instruments only


From   Rodolphe Desbordes <rodolphe.desbordes@strath.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: 2SLS with selected instruments only
Date   Mon, 8 Mar 2010 15:54:46 +0000

Dear Thu Phuong Pham,

What you are proposing is described by Angrist and Pischke ("Mostly Harmless Econometrics", 2009, p. 189) as one of the most common 2SLS mistakes. Your first stage must always include the same exogenous variables as those included in your second stage. If that is not the case, your estimator will be inconsistent.

Regarding your second question, I believe that the new version of -ivreg2- allows for arbitrary within-cluster correlation in two cluster
dimensions.

Best regards,

Rodolphe

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Thu Phuong Pham
Sent: lundi 8 mars 2010 06:27
To: statalist@hsphsun2.harvard.edu
Subject: st: 2SLS with selected instruments only

Hi

I would like to ask you about the command to run 2SLS with instruments that
can selected. I know that STATA provides IVREGRESS to run 2sls but I have
the following problem. For example, when I put this command:

IVREGRESS 2sls dependent control1 control 2 (control 3 = instrument 1
instrument 2);

STATA automatically treats CONTROL 1 and CONTROL 2 as INSTRUMENT 3 and
INSTRUMENT 4; while in my regression, CONTROL 1 and CONTROL 2 can NOT be
instruments for CONTROL 3 even they are exogenous. I want to regress CONTROL
3 on INSTRUMENT 1 and INSTRUMENT 2 only; but I do not know how I can do it.

Could you please advise me how to solve it?

Moreover I would like to calculate Thompson(2009) Standard Errors that
clusters both firms and time,  for OLS fixed effect and for this 2SLS; could
you please advise me how I can do that? Is there any option for it?

I am working on balanced panel data

Thank you very much for your help.

 Best regards,

 Thu Phuong Pham

*
*   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/

*
*   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