DOUBLE EXCEPT EXTREME RANKED SET SAMPLING FOR ESTIMATING POPULATION MEAN

Abstract

The ranked set sampling (RSS) method is an efficient sampling method that is used when the judgment ranking is easy while measuring the sampling units is expensive. This method has a higher efficiency than the commonly used simple random sampling (SRS) method. If this method is implemented in two stages, it will be called double-ranked set sampling (DRSS), which has a higher efficiency than RSS. Recently, the except extreme ranked set sampling (EERSS) method is proposed as a modification of the RSS. In the current study, we modified the EERSS and suggested the double-stage EERSS (DEERSS) method for estimating the population mean. The DEERSS estimator is compared with each DRSS, EERSS, RSS, and SRS counterparts. It has been shown that the DEERSS estimator is more efficient than all the competitor estimators considered in this study. Also, it is shown that this estimator is unbiased in the case of symmetrical distributions. Three data sets are employed to test the applicability and efficiency of the DEERSS method.