Efficient Class of Estimators of Population Variance Under Random Non-response in Two-Phase Successive Sampling

In this paper, an investigation has been carried out to deal with the problems of non-response in estimation of population variance on current occasion in two-occasion successive sampling under two-phase setup. Using readily available information on an auxiliary variable on both occasions, class of estimators of population variance on current occasion have been proposed in presence of random non-response on either of the occasions. Performances of the proposed class of estimators are compared with sample variance estimator under the similar realistic situations to study the properties of the suggested estimation procedures under missing at random response mechanism. Results are validated through empirical studies which justify the proposition of the class of estimators. Suitable recommendations have been made to the survey practitioners.