Relative Efficiency Of Product Estimation Strategies Under Super Population Models
by
Manoj Goyal
Directorate of Human Resource Management,CCS Haryana Agricultural University, Hisar, India-125004
Coauthors: K.C. Goyal, Veena Manocha

Singh and Singh (1997) investigated the relative efficiency of mean per unit estimator with conventional product estimator and the modified product estimators proposed by Robson (1957) and Dubey (1993) under Durbin’s super population model (1959). In this paper, it was observed that the estimators considered by them are inconsistent and that the product estimator is not efficient even than that of the mean per unit estimator whereas in sampling theory it is reported that the product estimator is always efficient than the mean per unit estimator if –2r > Cx/Cy , the ratio of coefficient of variation in variables X and Y. It is therefore concluded that the super population model considered by them is unsuitable to study the relative efficiency of product strategies. In the present article, an alternative super population model is proposed and the relative efficiency of product estimator with mean per unit and the regression estimator have been found. The situations when the product estimator is better than the regression estimator have been identified whereas the Srivastava (1980) found that the regression estimator can not be improved upon unto first order of approximation.

Date received: November 24, 2002

Copyright © 2002 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cakd-35.