Modified cut-off sampling is widely used for highly skewed data. A serious drawback of modified cut-off sampling is adjustment of non-response in take-all stratum. Therefore, the problems of non-response in take-all stratum have been studied in various way such as substitute of samples, imputation or re-weight method. In this paper, instead of considering imputation or re-weight method in take-all stratum, a sample design which can reduce the number of take-all stratum including non-response samples is suggested by minimizing. Simulation studies are performed and Labor Survey data are used for the case study.

Keywords: Non-response; Cut-off point; Skewness; MSE

Biography: Applied Statistics Professor in Kyonggi University