This preview shows page 1. Sign up to view the full content.
Unformatted text preview: ive-stain group was estimated to be 2.5 times greater than in the
negative-stain group, although this did not reach signiﬁcance (p = .0695). 14 Graphs to check whether hazards are really proportional Proportional hazards assumption: ratio of hazards is constant and does not
depend on time: h A (t )
h B (t ) When this assumption fails, it is because the hazard ratio changes over time.
Proc Lifetest makes 3 graphs that provide visual checks of this assumption: Proc Lifetest
plots=(SURVIVAL (atrisk = start to stop by step) LOGSURV
SURVIVAL plots estimated survivor functions.
If they cross, then hazard changes over time.
15 Stomach Cancer Breast Cancer 16 LOGLOGS ); LOGSURV plots the cumulative hazard function(s) H (t ) = − log S (t ) If hazards are proportional, then larger cumulative hazard should be a multiple of
smaller: Hi = r H j . Breast cancer example: 17 Stomach cancer example, where survivor curves crossed: 18
LOGLOGS gives a plot of log cumulative hazard log H (t ) = log − log S (t )
If hazards are proportional, then LOGLOGS plot will show parallel curves:
log Hi = c + log H j . Breast cancer example: 19 LOGLOGS for...
View Full Document
This note was uploaded on 11/21/2011 for the course PUBH 6470 taught by Professor Williamthomas during the Fall '11 term at University of Florida.
- Fall '11