Abstract:
|
Survival analysis provides insight into time-to-event data. Traditional techniques such as the Kaplan-Meier estimator, the log-rank test, and Cox regression mostly focus on right-censored data and are widely used in practice. By comparison, specialized methods for dealing with interval-censoring and competing risks are less well known, and practicing statisticians are often not aware that software is available for these methods. After reviewing basic concepts of survival analysis, this tutorial introduces two new procedures in SAS/STAT for the analysis of interval-censored data: PROC ICLIFETEST and PROC ICPHREG. The tutorial demonstrates how to use these procedures for estimation and comparison of survival functions and proportional hazards regression. The tutorial then turns to the analysis of competing risks data and explains how to use the LIFETEST procedure to conduct nonparametric survival analysis and the PHREG procedure to investigate the relationship of covariates to cause-specific failures. A basic understanding of applied statistics is assumed.
|