DOUBLE QUARTILE RANKED SET SAMPLING FOR ESTIMATING POPULATION RATIO USING AUXILIARY INFORMATION

In this paper, the problem of estimating the population ratio, R = mu(Y) / mu(X) using double quartile ranked set sampling (DQRSS) method is considered, where mu(Y) and mu(X) are the population means of Y (primary variable) and X (auxiliary variable) respectively. The DQRSS is compared with the ranked set sampling (RSS), quartile ranked set sampling (QRSS) and simple random sampling (SRS) methods based on the same sample size and correlation coefficient. The ranking is performed on both variables Y and X separately. A real data set is used to illustrate the method. It is found that the DQRSS estimators are approximately unbiased for the population ratio, and more efficient than their counterparts using SRS, RSS, and QRSS for all cases considered in this study. Also, it turns out that the ranking on the variable X is more efficient than the ranking on the variable Y.