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/
