An effective imputation method to minimize the effect of random non-response in estimation of population variance on successive occasions

In this paper, an attempt has been made to reduce the negative effect of random non-response in the estimation of population variance on the current occasion in two-occasion successive sampling. A difference type imputation method has been considered to minimize the nuisance effect of random non-response on both occasions. To build up efficient estimation strategies of population variance on current occasion intelligible use of auxiliary information has been made. Estimators are derived for the current occasion as special cases, when random non-response occurs either on the first or on the second occasion. To describe the effectiveness of the proposed estimators, empirical studies are carried out to compare their performances with the natural sample variance and ratio type estimator under the complete response situations. Results are interpreted through empirical studies, which are followed by suitable recommendations.