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/
