<[email protected]> points out an inconsistency in the estimation of the baseline
cumulative hazard using -stcox-.
> I would like to report what I believe is an erroneous result that is
> generated by Stata 7.
> After the estimation of a Cox proportional hazard model (stcox), I obtain a
> baseline cumulative hazard function that is NOT monotonically increasing
> with time. In particular, I obtain:
[...]
This is a bug, and it occurs only when one specifies the -efron- method for
handling ties. Also, as gc4 pointed out, the erroneous values occur at the
failure times for which there do exist ties.
We've added this bug to our things-to-do list.
--Bobby
[email protected]
*
* 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/
