Estimation Of Finite Population Mean Using Random Non–Response In Survey Sampling

This paper consider the problem of estimating the population mean under three different situations of random non–response envisaged by Singh et al (2000). Some ratio and product type estimators have been proposed and their properties are studied under an assumption that the number of sampling units on which information can not be obtained owing to random non–response follows some distribution. The suggested estimators are compared with the usual ratio and product estimators. An empirical study is carried out to show the performance of the suggested estimators over usual unbiased estimator, ratio and product estimators. A generalized version of the proposed ratio and product estimators is also given.