An e cient new class of estimators of population variance using information on auxiliary attribute in sample surveys

Abstract

This paper addresses the problem of estimating the population variance S 2 y of the study variate Y when information on an auxiliary attribute is available. We have suggested a new class of estimators for the population variance S 2 y using information on an auxiliary attribute. It has been shown that the proposed class of estimators is more ecient than those estimators considered by [16], [19] and [23]. The theoretical ndings are well supported through an empirical study.