Generalized regression cum ratio estimators of population variance in two phase sampling

This study suggested the generalized regression cum ratio estimators for estimating the population variance of the variable under study. The bias and mean squared error of the suggested estimator are derived under simple random sampling without replacement (SRSWOR) scheme. Empirical study has been performed with the three natural data sets which are having the dissimilar correlation among the study variable and the auxiliary variables. The proposed estimators are more efficient than the usual unbiased estimator, Isaki (1983) and Jhajj & Walia (2011) at all type of correlated populations.