An Estimator for the Survival Function on the Basis of Observable Censoring Times

• Published in: Communications in statistics. Theory and methods Vol. 32; no. 10; pp. 2043 - 2052
• Main Authors: Jensen, Uwe, Narr, Andreas
• Format: Journal Article
• Language: English
• Published: Philadelphia, PA Taylor & Francis Group 09-01-2003; Taylor & Francis
• Subjects: ACL-estimator; Exact sciences and technology; Informative random censoring; Koziol-Green model; Mathematics; Nonparametric inference; Nonparametric survival curve estimator; Parametric inference; Probability and statistics; Sciences and techniques of general use; Statistics; Censored data; Limit product estimator; Asymptotic behavior; Censored sample; Non parametric estimation; Statistical estimation; Observable; Estimator efficiency; Unbiased estimation; Survival function; Consistent estimator; Asymptotic approximation; Koziol Green model; 1999; Survival data
• ISSN: 0361-0926; 1532-415X
• DOI: 10.1081/STA-120023265

Chiu [Chiu, S. N. (
1999
). An unbiased estimator for the survival function of censored data. Commun. Statist. - Theory Meth. 28(9):2249-2260.] proposed a nonparametric estimator for the survival function which is based on observable censoring times in the general censoring model. His estimator is less efficient than the Product-Limit estimator. Considering an informative censoring model this drawback can partially be overcome. This is shown by a nonparametric, uniformly consistent estimator based on observable censoring times within the simple Koziol-Green model. Some asymptotic properties of the new estimator are investigated and it is compared with the well-known ACL-estimator.