AISM 54, 626-640
© 2002 ISM
(Received April 6, 2000; revised June 11, 2001)
Abstract. The ranked-set sampling (RSS) is applicable in practical problems where the variable of interest for an observed item is costly or time-consuming but the ranking of a set of items according to the variable can be easily done without actual measurement. In this article, the M-estimates of location parameters using RSS data are studied. We deal mainly with symmetric location families. The asymptotic properties of M-estimates based on ranked-set samples are established. The properties of unbalanced ranked-set sample M-estimates are employed to develop the methodology for determining optimal ranked-set sampling schemes. The asymptotic relative efficiencies of ranked-set sample M-estimates are investigated. Some simulation studies are reported.
Key words and phrases: Asymptotic normality, asymptotic relative efficiency, M-estimates, optimal sampling design, ranked-set sampling, robustness.
Source (TeX , DVI )