31-08-2013, 04:28 PM
A Class of Estimators of Population Ratio of Means in Presence of Non Response
Class of Estimators.docx (Size: 79.15 KB / Downloads: 21)
Abstract
In the presence of non response, a class of estimators of finite ratio of two population means is proposed when the population mean of auxiliary variable is not known; its bias and mean square error are found. Sub - Class of optimum estimators in the sense of having minimum
mean square error is found and enhancing the practical utility, a sub - classes of estimators depending on estimated optimum value based on sample observations is also investigated in the presence of non response. The expressions of sample size and inverse subsample fraction have been worked out by minimizing the cost for given MSE, and further the expressions of sample size and inverse subsample fraction have been found out by minimizing the MSE for fixed cost.
Introduction
In most of the sample surveys, the information cannot be obtained from all the units in
the survey. An estimator based on such incomplete information is generally biased and the
results may be grossly misleading when the respondents differ from the non-respondents. In
their seminal paper Hansen and Hurwitz (1946) considered a technique of sub-sampling the
non-respondents in order to adjust for the non-response bias in a mail survey.