Variance Estimation in Stratified Random Sampling in the Presence of Two Auxiliary Random Variables

The objective of this paper is to develop an improved population variance estimators in the presence of two auxiliary variables in stratified random sampling adapting the family of estimators proposed by Koyunchu and Kadilar (2009) for the estimation of population mean in stratified random sampling using prior information of the two auxiliary variables. In this paper, we proposed ratio-product type estimators and derived their mean square errors using first order approximation of Taylor series method. Efficiency comparisons of proposed estimators with respect to their mean square errors have been discussed and achieved improvement under certain conditions. Results are also supported by numerical analysis. Based on results, the proposed ratio- type variance estimators may be preferred over traditional ratio-type and sample estimator of population variance for the use in practical applications.