Hi Clint,
search stcoxgof on SSC archive, but it relies on a single record per subject.
I do not know if Hosmer and Lemeshow ever extended this approach to multiple records per subjects.
If you know this update, please send me reference.
Enzo
-----Messaggio originale-----
Da: [email protected] per conto di [email protected]
Inviato: ven 17/03/2006 16.51
A: [email protected]
Oggetto: st: -stcox- and Overall Goodness of Fit
Hello All --
This question involves the re-creation of a table intended to assess
goodness of fit following a cox proportional hazards estimation. More
specifically, I want to re-create the table provided on page 228 of Hosmer
& Lemeshow's "Applied Survival Analysis" (1999) wherein the table returns
the observed number of events, estimated number of events, z-scores, &
p-values w/in each decile of risk. Among a few problems, a major one I'm
having is the fact that my data is multi-record, that is, multiple records
constitute a single subject (read: an event history exists for each
subject) and I suspect that the proper creation of said table relies on
single-record data. Any thoughts? Is anyone aware of a user-written
command already in existence? I've searched the Stata help files and read
Stata's -stcox diagnostics- material, but to little avail...
Many Thanks in advance...
--Clint
*
* 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/
<<winmail.dat>>