st: predicting survival with a semiparameteric model

["Feiveson, Alan H. (JSC-SK311)" <alan.h.feiveson@nasa.gov>]

Hi - I am estimating  a semi-parametric survival model using -stcox-. In the process, I can get estimates of the cumulated baseline hazard (say H0(t)) evaluated at values of time, t, in my data. Then I can use this to predict the survival at time t for given values of the explanatory variables, say x, using


S(t) = exp(-exp(xb)H0(t))

where xb is the linear predictor.

However, I don't see how to get a standard error of this prediction. For example, -nlcom- only considers the estimation error in xb and treats H0(t) as a known constant.

So is there a way to incorporate the uncertainty in estimating H0(t) into the standard error of S(t)?

Of course, I can always get a standard error of S(t) with a fully parametric model using -streg-, but if possible, I'd like to use a PH model without having to specify a distribution.

Thanks

Al Feiveson

*
*   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/