Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | "Guiella Georges" <georges.guiella@umontreal.ca> |
| To | <statalist@hsphsun2.harvard.edu> |
| Subject | st: Competing risk in discrete time with stcrreg |
| Date | Fri, 26 Nov 2010 11:16:50 -0500 |
Dear listers, I am running an event-history (survival) analysis with discrete time (panel) data in the presence of competing events (according to Fine and Gray (1999)' definition): first sexual intercourse within union and first sexual intercourse outside a union. For this purpose, I am using the stcrreg command of STATA11. My question is the following: Does the stcrreg command allow me for modelling competing risks with discrete time panel data as you would do using "stplit" command for discrete time logit regression? If yes, how do you "stplit" your data? I looked at the pdf help documentation (page 211) where it is recommended that "you stset an ID variable that identifies the subjects" but I'm not sure that it is the procedure to enter in discrete time analysis. Thank you very much for your help. Georges Guiella Université de Montréal Tél: (514) 343-6111 Ext:1959 Canada * * 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/