Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: c-statistic in Cox models


From   "Newson, Roger B" <r.newson@imperial.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: c-statistic in Cox models
Date   Tue, 10 Oct 2006 16:10:56 +0100

I presume that the version of -somersd- that you are using is the old
Stata 6 version (also used by Stata 7 and Stata 8 users). The old Stata
6 version of -somersd- couldn't handle censored data correctly, as it
didn't have the -cenind()- option, which was introduced in the first
Stata 9 of -somersd- version in July 2005. The -estat concordance-
command was also introduced by StataCorp in the course of upgrading to
Stata 9. I seem to recall that -stcstat- was written in an even earlier
version of Stata, and didn't work even for Stata 8. Therefore, to
calculate Harrell's c for survival data, I think you need to update to
Stata 9, either by changing the policy of your workroom or otherwise.

I hope this helps.

Best wishes

Roger


Roger Newson
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: r.newson@imperial.ac.uk 
www.imperial.ac.uk/nhli/r.newson/

Opinions expressed are those of the author, not of the institution.

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ashwin
Ananthakrishnan
Sent: 10 October 2006 15:16
To: statalist@hsphsun2.harvard.edu
Subject: RE: st: c-statistic in Cox models

Thanks for the suggestion.

I installed the somersd package, but am not sure how
to interpret the results or what represents the
c-statistic. 

I input the dependent variable and the independent
variables, but it doesn't ask me for a specific time
I'm looking at (for example failure at 12 months or at
24 months). It outputs coefficients for each of the
variables I input, but doesn't give me a c-statistic
or overall fit. 

Am I not using it correctly?

I'm also not able to use the 'estat concordance'
command suggested by another user because it doesn't
seem to run on Stata 8.0 which is the version I have
at my workplace (which I cannot upgrade to 9.0 due to
some policy reasons in the workroom).

Thanks,
Ashwin

 


--- "Newson, Roger B" <r.newson@imperial.ac.uk> wrote:

> The -somersd- package, downloadable from SSC using
> the -ssc- command,,
> also calculates general Harrell c-indices, and also
> Somers' D and
> Kendall's tau-a. There is an example in the manual
> -somersd.pdf-, which
> is distributed with the package as an ancillary
> file.
> 
> I hope this helps.
> 
> Roger
> 
> 
> Roger Newson
> Lecturer in Medical Statistics
> Respiratory Epidemiology and Public Health Group
> National Heart and Lung Institute
> Imperial College London
> Royal Brompton campus
> Room 33, Emmanuel Kaye Building
> 1B Manresa Road
> London SW3 6LR
> UNITED KINGDOM
> Tel: +44 (0)20 7352 8121 ext 3381
> Fax: +44 (0)20 7351 8322
> Email: r.newson@imperial.ac.uk 
> www.imperial.ac.uk/nhli/r.newson/
> 
> Opinions expressed are those of the author, not of
> the institution.
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On
> Behalf Of Jacki Buros
> Sent: 08 October 2006 02:29
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: c-statistic in Cox models
> 
> Hi Ashwin,
> 
> Try the stcox postestimation command -estat
> concordance-, which
> calculates Harrell's C. This may be what you want
> for the survival
> analysis, so long as you do not have time-varying
> covariate effects.
> 
> I don't know whether there is a similar command for
> svy regress.
> 
> Jacki
> 
> On 10/7/06, Ashwin Ananthakrishnan
> <ashwinna@yahoo.com> wrote:
> > Hi,
> >
> > Is there a command in stata where I can calculate
> the
> > c-statistic for predictive models in Cox
> regression
> > survival analysis?
> >
> > I tried the stcstat program that I found online
> but
> > that doesn't seem to be working. Are there other
> > commands?
> >
> > Is there a similar c-statistic command for use in
> > survey analysis also? (following a .svylogit or a
> > .svyregress command) ?
> >
> > Thanks.
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam
> protection around
> > http://mail.yahoo.com
> > *
> > *   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/
> >
> *
> *   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/
> 
> *
> *   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/
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
*
*   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/

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index