A New Efficient Estimator of Finite Population Mean for Dealing Non-Response in Survey Sampling

Abstract

In this paper, we propose a new exponential product ratio estimator and an exponential regression estimator to estimate the mean value of a finite population of the study variable y using the auxiliary variable for dealing non-response. Using a large sample approximation, the expression of bias and mean error squared (MSE) are obtained for the proposed estimates. Under certain realistic conditions, the proposed estimation has been found to be more or equal accurate than normal unbiased estimates. The effectiveness of the proposed estimate is demonstrated by empirical studies in comparison with sample mean, ratio estimators and other current estimates.