Abstract: The paper deals with a new class of estimators of population mean under general sampling design whenever the problem of non-response exists in practice. Properties of the proposed class of estimators have been studied and an optimum class of estimators has been obtained. It is seen that the proposed class of estimators includes better estimators than usual regression type estimators. To support mathematical findings, numerical example is cited.
Keywords: Auxiliary information, non-response, bias, mean square error, relative efficiency, simple random sampling without replacement (SRSWOR)