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 at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: RE: Different coefficient magnitudes in ols and 2sls estimation


From   Tirthankar Chakravarty <tirthankar.chakravarty@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Different coefficient magnitudes in ols and 2sls estimation
Date   Fri, 23 Apr 2010 00:12:36 +0530

" Is this test focused on the two coefficients that were specified?"

Indeed. It is a quadratic in the difference of the estimators weighted
by the inverse of the difference of the variance-covariance matrices.

T

2010/4/23 Nick Cox <n.j.cox@durham.ac.uk>:
> I know nothing about this stuff. Is this test focused on the two coefficients that were specified? It sounds as if the reviewer were just saying "Why are these so different?", which could be a reasonable question.
>
> Nick
> n.j.cox@durham.ac.uk
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Tirthankar Chakravarty
> Sent: 22 April 2010 19:33
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: RE: Different coefficient magnitudes in ols and 2sls estimation
>
> I think, in this situation the Hausman test would be the formal
> procedure that tests the divergence between the two estimators. Under
> the null of Y2 being exogenous, both the OLS and 2SLS estimator are
> consistent and tend to the same probability limit, else the OLS is
> inconsistent and the probability limits of the two estimators diverge.
> The Hausman test is a formal tests of this convergence under the null.
>
> T
>
> 2010/4/22 Nick Cox <n.j.cox@durham.ac.uk>:
>> No, I meant whether your judgement "significantly smaller" reflected a
>> difference confirmed as such by an appropriate formal procedure. I don't
>> know what it would be, but this is not it.
>>
>> Nick
>> n.j.cox@durham.ac.uk
>>
>> Sukesh Patro
>> Sent: 22 April 2010 19:11
>> To: statalist@hsphsun2.harvard.edu
>> Subject: Re: st: RE: Different coefficient magnitudes in ols and 2sls
>> estimation
>>
>> Sorry Nick. I left out the t-stats. The OLS t-stat is about 3.00 and
>> the 2SLS t-stat is about 2.50. Is this what you meant? I only left
>> them out because they were part of the original question.
>>
>> Thanks again, SSKP
>>
>>
>> On Thu, Apr 22, 2010 at 1:07 PM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
>>> So, no formal significance test in either case. Thanks for the
>>> clarification.
>>>
>>> Nick
>>> n.j.cox@durham.ac.uk
>>
>> *
>> *   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/
>>
>
>
>
> --
> To every ω-consistent recursive class κ of formulae there correspond
> recursive class signs r, such that neither v Gen r nor Neg(v Gen r)
> belongs to Flg(κ) (where v is the free variable of r).
>
> *
> *   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/
>



-- 
To every ω-consistent recursive class κ of formulae there correspond
recursive class signs r, such that neither v Gen r nor Neg(v Gen r)
belongs to Flg(κ) (where v is the free variable of r).

*
*   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