Optimal Random Non Response Frameworks for Mean Estimation on Current Occasion.

In this paper, we propose various sampling strategies under random non response framework and proposed estimators to estimate population mean on current occasion. The formulation of the estimator is allied with Cochran (1977) and Searls (J. Am. Stat. Assoc.59, 1225–1226 1964) in the framework of random non response. The characteristics of each proposed estimator have been studied under optimum replacement policy. We have examined the performance of these estimators under the analytical study and validated it through numerical study. We have also gauged the loss with the available complete response case and reported in numerical illustration. Suitable recommendations have been put forward to the survey statisticians for its practical application. [ABSTRACT FROM AUTHOR]