Friday, March 17, 2006

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: * * *


<< Home

This page is powered by Blogger. Isn't yours?