Re: st: Residuals of a Cox regression?

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Thu, 9/12/10, Stata Chris wrote:
> I would like to plot the "residuals" of a Cox regression,
> but when I type -predict resid, residuals- after -stcox-
> Stata tells me that the option is not allowed here.

Have you looked at -help stcox postestimation-? Before 
adding an option to a command you must look at the 
appropriate help file to see if it is allowed and what
it does exactly. If you had looked at the helpfile you
would have seen that -residuals- is not a permitted option.

Residuals in nonlinear models are actually much more 
complicated to define than in a regular linear regression.
Moreover, -stcox- deliberately does not model a key 
aspect of the dependent variable: the baseline hazard. 
This is what makes Cox regression "semi-parametric", but
it also means that defining residuals is even harder. 
Several solutions have been proposed and implemented, but
the proper use and interpretation of these residuals is
far from straight forward. I suggest you do some reading
before trying these, e.g.:
<http://www.stata-press.com/books/saus3.html>

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------



      

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/