Hello,
I have a continuous time (counting-process) survival analysis with
multiple failures per person and multiple failure types per moment
time per person. So the Cox regression analysis is both stratified on
failure type and adjusted with either shared(personid) or
cluster(personid) -- though since one cannot specify both shared() and
strata() options with stcox for this combination I use GLLAMM.
I would now like to estimate the (Nelson-Aalen) cumulative failure
function at each failure time with either
(a) adjusted SE for clustering within persons
(b) shared frailty (person-level random effects)
with the goal of graphing the cumulative failure by time with CI. From
testing I believe the CI from 'sts graph, na cna' are not adjusted for
patient-level effects and so are inappropriate, but I could be wrong
as I don't have the [ST] manual.
Any advice as to how I might make either adjustment, perhaps using GLLAPRED?
Many thanks in advance,
Jacki
*
* 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/
