A New Family of Exponential Type Estimators in the Presence of Non-Response

Abstract

We propose families of estimators for the population mean using an exponential function in case of non-response. This situation is examined under two cases, Case I and II. The bias, MSE and minimum MSE are separately obtained for both cases. We compare the proposed estimators theoretically with the main estimators from the literature, such as Hansen and Hurwitz (1946), ratio, regression and exponential estimators. The conditions for which the proposed estimators are most efficient are obtained. Moreover, different empirical studies are conducted to support the theoretical results for both cases.

Keywords

• auxiliary variable
• efficiency
• exponential type family of estimators
• family of estimators
• non-response
• population mean