Statalist


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

RE: st: RE: testing for endogeneity with cluster()


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: testing for endogeneity with cluster()
Date   Thu, 25 Sep 2008 18:02:23 +0100

Rita,

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of 
> ritashree chakrabarti
> Sent: 25 September 2008 17:35
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: testing for endogeneity with cluster()
> 
> Hi Mark,
> 
> The example with the "ivreg2" help file shows that I have to use the
> "ivendog" command after I run the IV regression. This was what I was
> doing before and it gives the Durbin-Wu statistic. Is it the same as
> the endog() option that you are talking about which gives consistent
> even after clustering  the standard errors? I could not find the
> endog() option in my version of the ivreg2 help file.

Which version are you using?  From within Stata, use -which ivreg2-.
Mine shows:

. which ivreg2
c:\ado10\plus\i\ivreg2.ado
*! ivreg2 2.2.09  17jul2008
*! authors cfb & mes
*! see end of file for version comments

and the clickable example in the help file is:

. ivreg2 lwage exper expersq (educ=age kidslt6 kidsge6), endog(educ)

--Mark

> Thank you.
> 
> Rita.
> 
> On Thu, Sep 25, 2008 at 9:07 AM, ritashree chakrabarti
> <hirup01@gmail.com> wrote:
> > Hi Mark,
> >
> > Thanks for your help. So the latest version of "ivreg2" would take
> > into consideration that there is correlation between the 
> groups right?
> > I was thinking of testing for endogeneity by bootstrapping the
> > differences between the OLS and IV coefficients, which is a more
> > complicated process. The endog() would make things so much more
> > simpler.
> >
> > Ritashree
> >
> > On Wed, Sep 24, 2008 at 6:48 PM, Schaffer, Mark E 
> <M.E.Schaffer@hw.ac.uk> wrote:
> >> Ritashree,
> >>
> >>> ----d -Original Message-----
> >>> From: owner-statalist@hsphsun2.harvard.edu
> >>> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of
> >>> ritashree chakrabarti
> >>> Sent: 24 September 2008 17:14
> >>> To: statalist@hsphsun2.harvard.edu
> >>> Subject: st: testing for endogeneity with cluster()
> >>>
> >>> Hi,
> >>>
> >>> I am doing IV regressions and using the "ivreg2" command. 
> I want to
> >>> test for endogeneity and therefore using the "ivendog" to get the
> >>> Durbin-Wu-Hausman test statistic after conducting the IV 
> regressions.
> >>> However, I want to report results with robust standard 
> errors obtained
> >>> after clustering. Since this test is based on the 
> assumption that all
> >>> the observations are independent; the assumption will no longer be
> >>> valid when I pecify "cluster()" and, therefore, the test 
> produced by
> >>> -ivendog- is not valid in those cases, right.
> >>> My question is how to test for endogeneity when I am using the
> >>> "cluster()" with ivreg2?
> >>
> >> If you are using the latest version of -ivreg2-, just use 
> the -endog()-
> >> option.  There is a clickable example at the bottom of the 
> -ivreg2- help
> >> file.
> >>
> >>> Also to test for the validity of instruments after 
> clustering I should
> >>> be comparing the Hansen J-statistic reported 
> automatically when I run
> >>> "ivreg2" with the cluster option.
> >>
> >> I think the short answer is "yes".
> >>
> >> Best wishes,
> >> Mark
> >>
> >>> I am concerned whether I am taking
> >>> the right steps and interpreting the results correctly.
> >>>
> >>> I would really appreciate your help.
> >>>
> >>> Thanks.
> >>>
> >>> Ritashree.
> >>> *
> >>> *   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/
> >>
> >
> *
> *   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