On the Kaplan-Meier estimator based on ranked set samples Articles uri icon

publication date

  • December 2014

start page

  • 2577

end page

  • 2591

issue

  • 12

volume

  • 84

International Standard Serial Number (ISSN)

  • 0094-9655

Electronic International Standard Serial Number (EISSN)

  • 1563-5163

abstract

  • When quantification of all sampling units is expensive but a set of units can be ranked, without formal measurement, ranked set sampling (RSS) is a cost-efficient alternate to simple random sampling (SRS). In this paper, we study the Kaplan&-Meier estimator of survival probability based on RSS under random censoring time setup, and propose nonparametric estimators of the population mean. We present a simulation study to compare the performance of the suggested estimators. It turns out that RSS design can yield a substantial improvement in efficiency over the SRS design. Additionally, we apply the proposed methods to a real data set from an environmental study.

keywords

  • ranked set sampling; censored data; kaplan–meier