help with a survplot

On Sun, May 1, 2011 at 4:49 AM, David Winsemius <dwinsemius at comcast.net 

wrote:

On Apr 30, 2011, at 10:44 AM, Jabba wrote:

Dear useRs,

I was asked to produce a survival curve like this:

http://www.palug.net/Members/jabba/immaginetta.png/view

with the cardinality of the riskset at the bottom.

The 2nd sentence of help page for survival::survplot says that is  
an inbuilt
option.

There isn't a survival::survplot.  Perhaps you mean rms::survplot?

Apologies. I was confused about which package it was in. I generally  
load survival by require()-ing rms, and thought (without looking at  
what was plainly to see) was from survival.

    -thomas

-- 
Thomas Lumley
Professor of Biostatistics
University of Auckland

David Winsemius, MD
West Hartford, CT

Thank you very much. 

Despite prof. Harrell's support (for whom I feel great
esteem) I still remain doubtful about this feature.

Please elaborate.
Thanks
Frank

Thank you very much. 

Despite prof. Harrell's support (for whom I feel great
esteem) I still remain doubtful about this feature.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

-----
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context: http://r.789695.n4.nabble.com/help-with-a-survplot-tp3485998p3490126.html
Sent from the R help mailing list archive at Nabble.com.

Il giorno Mon, 2 May 2011 07:30:10 -0700 (PDT)
Frank Harrell <f.harrell at vanderbilt.edu> ha scritto:

Please elaborate.

It is simply that, generally speaking, i don't like adding numbers to a
plot. I eventually realized that riskset cardinality may be a useful
indication. However, do not discriminate between events and censored
observations could generate some confusion, i think.

Anyway, thanks to your software i was able to produce what i was
requested for in very little time, so thank you very much.

Hi Marco,

You're welcome.

The number at risk at given time points is a fairly standard thing to add to
survival plots.  I don't think many people will confuse it with numbers of
events.  Personally I find shaded confidence bands a bit more helpful but
both are useful.

Frank

Il giorno Mon, 2 May 2011 07:30:10 -0700 (PDT)
Frank Harrell <f.harrell at vanderbilt.edu> ha scritto:

Please elaborate.

It is simply that, generally speaking, i don't like adding numbers to a
plot. I eventually realized that riskset cardinality may be a useful
indication. However, do not discriminate between events and censored
observations could generate some confusion, i think.

Anyway, thanks to your software i was able to produce what i was
requested for in very little time, so thank you very much.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

-----
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context: http://r.789695.n4.nabble.com/help-with-a-survplot-tp3485998p3500488.html
Sent from the R help mailing list archive at Nabble.com.

Il giorno Thu, 5 May 2011 18:42:11 -0700 (PDT)
Frank Harrell <f.harrell at vanderbilt.edu> ha scritto:

Hi Marco,

You're welcome.

The number at risk at given time points is a fairly standard thing to
add to survival plots. 

I know, but last year, as a "newbye" in biostatistics, i felt the need
to read rms book exactly because there were plenty of "standard things"
that did not convince me.