Efficient Family of Predictors of Unknown Population Median under Sampling Design  
Author H. S. Jhajj


Co-Author(s) Harpreet Kaur


Abstract For predicting the population median of variable under study, a family of efficient predictors has been defined by using prior information of population parameters based upon concomitant variables under sampling design. The comparison of proposed family of predictors has been made with the existing ones with respect to their mean square errors. It has been shown that efficient predictors can be developed from the family under the given practical situations which will have smaller mean square error than the best considered linear regression type predictor. Effort has been made to illustrate the results numerically as well as graphically by taking some empirical population considered in the literature.


Keywords Concomitant variables; Bias; Mean square error; Median; Two-phase sampling
    Article #:  20238
Proceedings of the 20th ISSAT International Conference on Reliability and Quality in Design
August 7-9, 2014 - Seattle, Washington, U.S.A.