Simulation study and implementation of the tests based on weighted Turnbull's estimators for interval-censored data

Several tests for comparing k groups of interval-censored data based on Tur nbull's estimator of a survival function are compared by a Monte Carlo simu lation study. The tests under consideration include the IWD test proposed b y Petroni and Wolfe, the score test derived by Finkelstein, and a logrank t ype test suggested by Sun. With a large sample size and in a two-sample com parison, simulation results suggest that the IWD test is applicable for ear ly and crossing hazard difference alternatives, whereas Finkelstein's score test. Sun's test and the logrank test are slightly more powerful than the IWD tests under late hazard difference alternatives. The IWD test is design ed for the two-sample testing problem. With a large sample size and in a k- sample comparison, Finkelstein's score test is slightly better than the oth er tests under a proportional hazards model, while Sun's test and the logra nk test are slightly better than the score test under early and crossing ha zard difference alternatives. However, none of the test statistics is unifo rmly better than the others for the alternatives considered in this paper w ith both non-overlapping and overlapping interval-censored data. Copyright (C) 2001 John Wiley & Sons, Ltd.