Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Re: System 2sls with robust standard errors


From   "Rodrigo A. Alfaro" <[email protected]>
To   <[email protected]>
Subject   Re: st: Re: System 2sls with robust standard errors
Date   Tue, 29 Aug 2006 22:05:31 -0400

Belen, 

I am almost sure that:

reg3 (y1 x1 x2 x3 y2)(y2 z1 z2 z3), 2sls

is equivalent to: 

reg y1 x1 x2 x3 y2 (z1 z2 z3)
reg y2 z1 z2 z3

.. right? You can check it.

Therefore a robust version is

reg y1 x1 x2 x3 y2 (z1 z2 z3), robust
reg y2 z1 z2 z3, robust

I don't think that robust-issue is controlled
by this because y1 does not affect y2. 

Rodrigo.


----- Original Message ----- 
From: "Belen Dias" <[email protected]>
To: <[email protected]>
Sent: Tuesday, August 29, 2006 12:53 PM
Subject: Re: st: Re: System 2sls with robust standard errors


Hi Rodrigo,
thank you for your advice, I think is may be very
helpful. The thing is, however, that I still am not
able to use -_robust- properly, as I understand very
little of the programmer's language in Stata. If you
can walk me through it, I would really appreciate it.

The first step would be this one

-reg3 (y1 x1 x2 x3 y2)(y2 z1 z2 z3), 2sls

Then, to predict the residuals, 

-predict r1, res eq(#1)
-predict r2, res eq(#2)

Then, I should be able to use -_robust- in the
residuals, but even if I don't get an error message, I
am not able to see the results. 

Thanks a lot

Belen





Belen,

If you can replicate the coefficients of the paper 
with -reg3-, use -_robust- command after that. 

Rodrigo.


----- Original Message ----- 
From: "Belen Dias" <[email protected]>
To: <[email protected]>
Sent: Friday, August 25, 2006 2:01 PM
Subject: st: System 2sls with robust standard errors


Hi,I want to replicate Alesina and Perotti's (1996)
result on incomedistribution, investment and political
instability. I notice that by the end of the paper
they use heteroscedasticity-consistent standard errors
with system 2sls. Now, I know that it is impossible to
get robust SE with the -reg3- command. 

Does anybody know if there is another way of doing
this? Can I do it "by hand"? (If so, how? Just doing
the 2sls "by hand" with -reg, r-?) 

Otherwise, it would be helpful if you could advise me
of other statistical packages that could handle this
issue.

Please send me all your insights for this issue.
Thanks in advance!
Belen

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam
protection around 
http://mail.yahoo.com 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index