Study of imputation based generalized classes of dual to product cum dual to ratio estimators for missing data in two-phase sampling

Abstract

Present paper proposes three generalized classes of estimators for estimating population mean under the framework of two-phase sampling design in the presence of non-response in the study variable by using auxiliary information and also the expressions for bias and mean square error are derived. Further, theoretical results stating superiority of the proposed estimators, over the existing estimators by an empirical study based on different data sets from the classical statistical literature are shown.