Improvement in Estimating the Population Mean in Double Extreme Ranked Set Sampling

In this paper, modified ratio estimators of the population mean are suggested using double extreme ranked set sampling (DERSS) method. The newly suggested estimators are compared with their counterparts suggested by Singh and Tailor (2003) using simple random sampling (SRS), and with extreme ranked set sampling (ERSS) methods. It is shown that, the DERSS estimators are approximately unbiased of the population mean. It is found that, DERSS is more efficient than SRS and ERSS for estimating the population mean based on the same num-ber of measured units. Also, the mean square error (MSE) of the DERSS ratio estimators with negative values of the correlation coefficient is less than the MSE with positive values