GENERALIZED FAMILY OF ESTIMATORS IN STRATIFIED RANDOM SAMPLING USING SUBSAMPLING OF NON-RESPONDENTS
Generalized family of estimators have been proposed for the estimation of population mean using two auxiliary variables in stratified random sampling under non-response. The bias and the mean square error expressions of the adapted and suggested family estimators have been derived up to first order of approximation. The mathematical conditions have been obtained for which the suggested family of estimators are more efficient than the stratified mean estimator and ratio estimator. Empirical study is also provided in support of theoretical findings. In addition, the expressions for optimum sample size of the each stratum in respect to cost of the survey have also obtained.