Abstract

Experiments, for example, medical and social science surveys, conducted by human are often characterized by problem of non-response of missing observations. In this study, an alternative imputation method to Singh and Horn (2000) and Singh et al (2014) compromised imputation methods has been suggested due to decrease in their efficiency when the value of unknown weights tend to unity. The properties (bias and MSE) of the proposed estimators were derived up to first order approximation using Taylor series approach. Conditions for which the proposed estimator more efficient than other estimators considered in the study were also established. Numerical illustration was conducted and the results revealed that the proposed estimator is more efficient.

Key words: Imputation, Non-response, Estimator, Population Mean, Mean Squared Error (MSE)