R help - Apr 2001 - standard errors of fitted values are different S-plus survival pa ckage and R

Perhaps this question has been asked before: but using the function predict(
fit,type="terms",se.fit=T), where fit is a coxph object in S-plus, the
estimated standard errors are different.  It may be different estimators of
the variance of the residuals?  Which one is the default in R, I don't find
that too easily in the documentation.  Does anybody know?

I'll be very grateful if somebody can help in this matter.

Georgina Bermann
Biostatistics, AstraZeneca R&D M?lndal
----
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

On Mon, 2 Apr 2001 Georgina.Bermann at astrazeneca.com wrote:
> Perhaps this question has been asked before: but using the function
predict(
> fit,type="terms",se.fit=T), where fit is a coxph object in
S-plus, the
> estimated standard errors are different.  It may be different estimators of
> the variance of the residuals?  Which one is the default in R, I don't
find
> that too easily in the documentation.  Does anybody know?
Which version of S-PLUS?  Can you give an example?  Issues include the
setting of the contrasts options, and that survival5 in R is later than
that in most S-PLUS versions (e.g. except 6.0 post-Linux versions).


-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

On Mon, 2 Apr 2001 Georgina.Bermann at astrazeneca.com wrote:
> Perhaps this question has been asked before: but using the function
predict(
> fit,type="terms",se.fit=T), where fit is a coxph object in
S-plus, the
> estimated standard errors are different.  It may be different estimators of
> the variance of the residuals?  Which one is the default in R, I don't
find
> that too easily in the documentation.  Does anybody know?
The variance of the residuals doesn't come into it for coxph.
predict.coxph(terms, se.fit=TRUE) was wrong and is now fixed. I'll send
you a revised version.

	-thomas

Thomas Lumley			Asst. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._