Back To Index Previous Article Next Article Full Text


Statistica Sinica 12(2002), 1073-1083



NONPARAMETRIC SURVIVAL COMPARISONS FOR

INTERVAL-CENSORED CONTINUOUS DATA


Hong-Bin Fang, Jianguo Sun and Mei-Ling Ting Lee


St. Jude Children's Research Hospital, University of Missouri-Columbia

and Harvard Medical School


Abstract: The problem of two-sample survival comparisons has been investigated by several authors. Pepe and Fleming (1989) introduced a test for right-censored survival data. Petroni and Wolfe (1994) considered a similar test where the survival times can only take on a finite number of values. We extend their tests to general types of interval-censored data and introduce a class of test statistics based on integrated weighted differences between the two estimated survival functions.We derive the asymptotic distributions of the generalized test statistics, present a bootstrap test procedure and apply the proposed method to a set of interval-censored data from a breast cancer study.



Key words and phrases: Breast cancer, bootstrap method, continuous survival time, interval-censored data, two-sample comparison.



Back To Index Previous Article Next Article Full Text