An Idea Exchange for SAS software and services

Comments
by Occasional Contributor GZ
on ‎03-28-2016 06:50 PM
I really hope the DDFM method can be added to PHREG in next release. Very frequently we make inference about the fixed effects at multi-levels using survival data which are of small sample size or unbalanced among clusters. Without appropriate adjustment for the denominator degrees of freedom, the results (i.e. the default p values) can just be invalid. STATA 14 has done this with its MESTREG command. SAS may implement an addition of the feature directly? I look forward to seeing and using the new function in survival analysis.
by Occasional Contributor GZ
on ‎05-09-2016 01:04 AM

STATA 14 MESTREG command may actually have no such functionality yet.   STATA web site (and Timberlake site) had a page titled “Multilevel survival models and Small-sample inference”, which could easily get readers confused. Nevertheless, SAS Institute is able and may well be the first to add this very useful feature to its product.

Idea Statuses
Top Liked Authors