st: Presenting variation in risk over time using stcrreg

["Travis Coan" <tcoan@law.harvard.edu>]

Dear Statalist,

I'm new to the Fine and Gray (1999) approach implemented by -stcrreg-
and I have a general question associated with presenting the risk or
hazard of failure over time after the command. 

Stata's -stcurve- command does not allow one to plot the subhazard
function after -stcrreg- (probably for good reason, though I'm still
trying to work through why something similar to Cox regression could not
be offered). Does anyone have suggestions for effective ways to present
variation in risk over time in the competing risk situation captured by
-stcrreg- (that is, one that actually takes competing events into
account)? Is there something similar to plotting the hazard function in
-stcox- for -stcrreg-?

I've considered possibly smoothing the subhazard contributions or just
focusing on changes in the probability using the CIF; however, I may be
way off here. I'd be interested to hear if anyone has thoughts on the
matter.

Thanks for your consideration!

Cheers,
Travis




*
*   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/