Simultaneous Confidence Bands from Two-Sample Censored Data

• Activity Number: 604
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Sundarraman Subramanian*+ and Nubyra Ahmed
• Companies: New Jersey Institute of Technology and New Jersey Institute of Technology
• Keywords: Bootstrap process ; Empirical coverage probability ; Estimated average enclosed area ; Maximum likelihood estimator ; Semiparametric random censorship models ; Survival function estimator

Abstract

The standard censored data bootstrap is employed to obtain nonparametric simultaneous confidence bands for the difference of two survival curves. Furthermore, a recently developed two-stage bootstrap is exploited to obtain semiparametric confidence bands for the difference. The two-stage bootstrap combines classical bootstrap with a model-based regeneration of censoring indicators. Simulation studies are presented to show that both types of proposed confidence bands are superior to a currently existing one, in the sense of having empirical coverage closer to the nominal level. Further comparison between the proposed confidence bands reveal that the model-based ones are superior, tighter and, hence, more informative. The methods are illustrated using real data sets.