Re: st: R: Estimating the probability of censoring

[Michael McCulloch <mm@pinest.org>]

Antoine Terracol made the brilliant suggestion of estimating 
probability of censoring, based on -stcox- with specifying the 
baseline hazard option basechazard(). In studying the manual (v9, 
page 147) section on the Technical Note for Example 7 of -stcox-, I 
read that specifying the baseline hazard option requires all 
variables to make sense if they would be equal to zero (which of 
course would not be the case with a variable such as age). That makes 
sense to me.

At the bottom of that page, I read that this can be resolved by 
subtracting mean age from each subject's age. Is this what is 
referred to as "centering on age"?
Michael



> Antoine Terracol wrote:
> > Michael,
> >
> > What is calculated here is the survival function at time t 
> > conditional on covariates, i.e., in your case, the probability that 
> > the subjects time-to-censoring will be greater than t if dying from 
> > cancer was not a risk.
> Of course, this is because Michael specified
>
> . stset studytime, failure(died==0)
>
> i.e. he is modelling time-to-censoring instead of the usual time-to-failure.
>
> Antoine
> *
> *   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/
*
*   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/