At 00:56 09/05/2008, Ted Harding wrote:
I'd like to thank Paul Johnson and Achim Zeileis heartily
for their thorough and accurate responses to my query.
I think that the details og how to use the procedure, and
of its variants, which they have sent to the list should
be definitive -- and very he
I'd like to thank Paul Johnson and Achim Zeileis heartily
for their thorough and accurate responses to my query.
I think that the details og how to use the procedure, and
of its variants, which they have sent to the list should
be definitive -- and very helpfully usable -- for folks
like myself wh
Once again, Paul, many thanks for your thorough examination
of this question! And for spelling out your approach!!!
It certainly looks as though you're very close to target
(or even spot-on).
I've only one comment -- see at end.
On 08-May-08 20:35:38, Paul Johnson wrote:
> Ted Harding said:
>> I
Paul & Ted:
> > I can get the estimated RRs from
>
> > RRs <- exp(summary(GLM)$coef[,1])
>
> > but do not see how to implement confidence intervals based
> > on "robust error variances" using the output in GLM.
>
>
> Thanks for the link to the data. Here's my best guess. If you use
> the follow
Ted Harding said:
> I can get the estimated RRs from
> RRs <- exp(summary(GLM)$coef[,1])
> but do not see how to implement confidence intervals based
> on "robust error variances" using the output in GLM.
Thanks for the link to the data. Here's my best guess. If you use
the following approac
On Thu, May 8, 2008 at 8:38 AM, Ted Harding
<[EMAIL PROTECTED]> wrote:
> The below is an old thread:
>
> On 02-Jun-04 10:52:29, Lutz Ph. Breitling wrote:
> > Dear all,
> >
> > i am trying to redo the 'eyestudy' analysis presented on the site
> > http://www.ats.ucla.edu/stat/stata/faq/relative_
The below is an old thread:
On 02-Jun-04 10:52:29, Lutz Ph. Breitling wrote:
> Dear all,
>
> i am trying to redo the 'eyestudy' analysis presented on the site
> http://www.ats.ucla.edu/stat/stata/faq/relative_risk.htm
> with R (1.9.0), with special interest in the section on "relative
> risk esti
7 matches
Mail list logo
