Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: RE: Heteroscedasticity-robust SEs in fixed effects panel.


From   Nick Cox <[email protected]>
To   [email protected]
Subject   Re: st: RE: Heteroscedasticity-robust SEs in fixed effects panel.
Date   Fri, 4 May 2012 10:05:16 +0100

-xtivreg2- is a user-written program from SSC. It requires only Stata
8.2. What leads you to suppose otherwise?

Nick

On Fri, May 4, 2012 at 9:50 AM, christina sakali
<[email protected]> wrote:
> Mark,
>
> The problem is I am using Stata 9 and as far as I know the choice you
> describe can only be found in newer versions of Stata. Is this right?
> But thanks anyway, knowing there are alternatives is still a big help.
>
> On 4 May 2012 03:26, Schaffer, Mark E <[email protected]> wrote:
>> Christina,
>>
>> -xtivreg2- has an undocumented option -sw- that will cause it to report
>> Stock-Watson heteroskedasticity-robust SEs for the fixed effects
>> estimator as described in their 2008 paper.  It's undocumented because I
>> haven't (yet) found a published or other set of results that would let
>> me confirm the coding with a replication.  I _think_ it's right ... but
>> caveat emptor.
>>
>> --Mark
>>
>>> -----Original Message-----
>>> From: [email protected]
>>> [mailto:[email protected]] On Behalf Of
>>> christina sakali
>>> Sent: 04 May 2012 00:45
>>> To: statalist
>>> Subject: st: Heteroscedasticity-robust SEs in fixed effects panel.
>>>
>>> Dear Statalist users,
>>>
>>> I am estimating a fixed effects panel regression with only 70
>>> observations (14 cross-sections, 5 years).
>>>
>>> I am wondering if anyone can suggest ways to obtain
>>> heteroscedasticity-robust S.E.s apart from the standard - fe
>>> robust - approach.
>>>
>>> The reason I am asking is that I am aware of Stock & Watson's
>>> (2008) findings about the bias in the standard het-robust SEs
>>> (Huber-White SEs), especially in the case of small T large N samples.

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index