In estimating population means for a study variable from random sampling, it is often possible to reduce the bias and improve the efficiency of estimators by including known data on an auxiliary variable which is correlated with the study variable. Non-response is a common problem that occurs when estimating values of variables by random sampling as it can increase the bias and reduce the efficiency of estimators. In this paper, we propose a new family of estimators for population means of a study variable when non-response occurs in the sampling of the study variable and when information on an auxiliary variable is either known or can be obtained by non-response sampling. The asymptotic properties of the proposed estimators such as bias, mean 2 square error (MSE), and minimum mean square error have been derived up to a first order approximation. A numerical study of the new estimators shows that they are more efficient than other existing estimators.

Abstract. This paper presents a new general family of estimators to estimate the population mean of study variable y in the presence of non-response when utilizing a known coefficient of variation of study variable y. The expressions for bias, mean squared error (MSE), and minimum mean squared error (MMSE) of the proposed family of estimators are derived up to the first degree of approximation. In addition, a numerical example and a simulation study are presented to explain the performance of the proposed estimators. It was shown that the proposed estimators perform better compared to all other relevant estimators.

No abstract

No abstract