Allin,
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Allin Cottrell
> Sent: 03 August 2006 01:56
> To: [email protected]
> Subject: Re: st: Newey estimations
>
> On Thu, 3 Aug 2006, Clive Nicholas wrote:
>
> >> Does somebody knows what is the exactly procedure Stata follows to
> >> test parameters under linear reg. with Newey-West std. errors
> >> estimators?
> >
> > Why use a wimpish, water-pistol version of OLS Newey-West in Stata
> > when you can use a OLS Newey-West package that's chock-full with
> > AK47s, Desert Eagles and Kalashnikovs? An
> > example:
>
> <snip>
>
> > Total (centered) SS = 9359943.917
>
> These 10-digit, "smart-bomb" results are verified using
> multiple-precision arithmetic, I trust?
>
> Allin Cottrell
> Wake Forest University
Actually, this bit is also part of the water pistol. Stata's -newey-
calls the built-in workhorse -regress-, and although the former doesn't
report the SS, the latter does calculate it:
. webuse grunfeld
. regress invest mvalue kstock
<snip>
Total | 9359943.92 199 47034.8941
--Mark
*
* 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/
