Accelerated Failure Time Models
Competing risks analysis with accelerated failure time models is basically the same as with Cox models. As before, the key point is to treat all events as censoring except the one that you’re focusing on. However, there are some complications that arise in constructing tests of equality of coefficients across event types. There are also some special characteristics of the leader data that require slightly different treatment with AFT models.
One of those characteristics is the presence of zeros in the YEARS variable. Recall that any leaders who served less than one year were assigned a time value of 0. Of the 472 leaders, 172 fell in this category. This poses no problem for PROC PHREG, which is only concerned ...
Get Survival Analysis Using SAS®: A Practical Guide now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
