Re: st: Numerous t-tests

["Michael Blasnik" <michael.blasnik@verizon.net>]

That's true.  My point was only that the estimated uncertainties from 
regress may not match those from ttest when unequal variances are 
appropriate.  I believe that's true whether you use the robust option for 
regresss or not.

Michael Blasnik
michael.blasnik@verizon.net


----- Original Message ----- 
From: "Richard Goldstein" <richgold@ix.netcom.com>
To: <statalist@hsphsun2.harvard.edu>
Sent: Wednesday, October 26, 2005 12:06 PM
Subject: Re: st: Numerous t-tests



> while I agree with Michael on this, note that using the robust
> option does a great deal to mitigate this
>
> Rich
>
> Michael Blasnik wrote:
> > "Ron�n Conroy" <rconroy@rcsi.ie> wrote:
> >
> > > -regress- is a lot faster than t-test, and the output is easier to 
> > > read, IMO. The coefficient and its confidence interval represent the 
> > > difference in means and its CI.
> >
> > While I agree that regress is faster and often preferred to t-tests, one 
> > should be aware that regress will assume equal variances for the two 
> > groups identified by the dummy variable.  That assumption may or may not 
> > be what you want.
> >
> > Michael Blasnik
> > michael.blasnik@verizon.net
*
*   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/