Statalist


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

RE: st: Stata 11 Random Effects--Std. Errors


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Stata 11 Random Effects--Std. Errors
Date   Tue, 18 Aug 2009 23:11:57 +0100

Jeff, Steve et al.,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> Jeff Pitblado, StataCorp LP
> Sent: 18 August 2009 22:53
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: Stata 11 Random Effects--Std. Errors
> 
> Steven Archambault <archstevej@gmail.com> is getting 
> different robust standard errors in Stata 11 than in Stata 10:
> 
> > I am absolutely getting different Robust std. errors in the 
> results of 
> > my newly downloaded Stata 11 than my old Stata 10.  This 
> occurs only 
> > for Robust Std. Errors in a Random Effects regression.  I 
> loaded both 
> > versions of the program and ran the same regression side by 
> side. This 
> > has an impact on my hausman test and the xtoverid test. I 
> cannot seem 
> > to find out what the difference in calculation methods are 
> between 10 
> > and 11.
> > 
> > Any insight would be helpful.
> 
> Steven hasn't shown us what estimation command he is using to 
> fit his random effects regression, but I suspect he is using 
> -ivreg2- or -xtivreg2- given that he also directly contacted 
> Mark Schaffer in his post to Statalist.

I think Steve is using official -xtreg- with random effects plus
-robust-.

The SEs under Stata 10 are different from those under Stata 11.  I think
this is because under Stata 10, -re- plus -robust- did not imply
-cluster-, but under Stata 11, -re- plus -robust- automatically means
-cluster-.

The manual is kind of opaque on why this change in the behaviour of
-xtreg- has been introduced, so if Jeff or anyone else can provide more
details, I and I'm sure others would be very interested.

Cheers,
Mark

> 
> Last Friday there was an ado-file update for Stata 11 that 
> fixed a problem with weights and the -_robust- command.
> 
> Steven should type the following in Stata
> 
> 	. update query
> 
> then follow the recommendation, which will be to 
> 
> 	. update all
> 
> if his Stata isn't already fully up-to-date.
> 
> This fix for -_robust- should resolve the differences that 
> Steven is currently observing.
> 
> Let me re-iterate from my original post, the bug in -_robust- 
> (fixed in last Friday's ado-file update) affeted only 
> user-written commands that used
> -_robust- directly; it did not affect any of Stata's official 
> estimation commands or any user-written estimators that used 
> -ml- or Mata functions
> -optimize()- or -moptimize()-.
> 
> Steven mentioned that he is not using weights, so I will be 
> contacting him directly to get to the bottom of this.
> 
> --Jeff
> jpitblado@stata.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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index