Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: further question about competing risks using Lunn & McNeil method


From   natalie chan <natalie.channew@googlemail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: further question about competing risks using Lunn & McNeil method
Date   Sun, 5 Apr 2009 22:56:57 +0100

Dear Enzo, Phil and other statalisters,

I've got another question about competing risks in Stata and would be
grateful if anyone could help.

I am trying to use Lunn & McNeil (1995) method. My dataset has about
3500 subjects and 30000 observations, each subject having multiple
records.

I expanded the data according to the method described in Lunn &
McNeil. But, when I was trying to -stset-, there is a problem. Seems
-stset- only allows one record for one id-time entry, the duplicated
rows are recognised as "overlapping records".

Here is the message from Stata output:

  30567  total obs.
      61134  overlapping records (time1[_n-1]>time0)              PROBABLE ERROR


Does anyone have any solution to this?

Many thanks.

Best

Natalie
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index