“School of Mathematics”

Back to Papers Home
Back to Papers of School of Mathematics

Paper   IPM / M / 16699
School of Mathematics
  Title:   New estimator for the variances of strata in ranked set sampling
  Author(s):  Ehsan Zamanzade
  Status:   Published
  Journal: Soft Computing
  Vol.:  25
  Year:  2021
  Pages:   8007-8013
  Supported by:  IPM
  Abstract:
Ranked set sampling (RSS) utilizes imprecise rankings on the variable of interest in order to draw an informative sample from the target population. The resulting sample, consisting of independent judgment order statistics, resembles a stratified random sample. Estimating the variances of strata is an important problem in RSS. The standard method is based on the sample variance of units in each stratum. A plug-in estimator is also available in the literature that remedies some shortcomings of the standard estimator. We adjust the latter estimator using kernel estimator of the distribution function. The developed estimator is shown to be consistent, and its performance is investigated by means of simulation. It turns out that our proposal can be considerably more efficient than the existing estimators when perfect or nearly perfect ranking holds.

Download TeX format
back to top
scroll left or right