Medyan Sıralı Küme Örneklemesi Yönteminde Kitle Ortalamasının Tahmini

Abstract

Ranked set sampling methods are used when it is costly or difficult to measure the sampling units precisely with respect to the variable of interest and also when these units can be more easily ranked with respect to the variable of interest or the auxiliary variable correlated with that variable of interest. In this study classic, ratio, regression, regression-ratio estimators and exponential family of estimators of population mean are examined through ranked set sampling, median ranked set sampling and simple random sampling methods. The efficiency of the ranked set sampling methods is directly related to the correct ordering. It is also related to the variable used in the ordering. Thus, within the scope of this study the results are examined through ranking both the variable of interest and auxillary variable. The estimators are compared in terms of relative efficiency by calculating the mean square error of the estimators through simulation study by using the air quality index data set. According to the simulation study, the most efficient results are obtained through the median ranked set sampling method.