I would like to use stcrreg's postestimation predict,basecif to calculate an annual probability of an event across several subjects.
At this point I'm not interested in any covariates but have approximately 5 competing risks. My subjects are followed across several years, but at this point the data consists of a single row per subject. Is the only way to stsplit my data, (i.e. divide each subject's record into a number of rows each representing a single year) in order to calculate an annual CIF?
My first attempt, without splitting the data, resulted in a number I'm not really sure I can interpret.
Thanks,
HR
Hugh S Robinson Ph.D.
Postdoctoral Researcher
Montana Cooperative Wildlife
Research Unit
University of Montana
Missoula, MT 59812
Ph. (406) 243-2476
*
* 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/
