Modified ratio estimators using stratified ranked set sampling

Abstract

Stratified Ranked Set Sampling (SRSS) combines the advantages of stratification and Ranked set sampling (RSS) to obtain an unbiased estimator for the population mean, with potentially significant gains in efficiency. The present paper deals with modified ratio estimators of finite population mean using information on coefficient of variation and co-efficient of kurtosis of auxiliary variable in Stratified Ranked Set Sampling. It has been shown that these methods are highly beneficial to the estimation based on Stratified Simple Random Sampling (SSRS). The bias and mean squared error of the proposed estimators with large sample approximation are derived. Theoretically, it is shown that these suggested estimators are asymptotically more efficient than the estimators in stratified simple random sampling. The results have been illustrated by numerical example.