Sankhya: The Indian Journal of Statistics

1994, Volume 56, Series B, Pt. 3 , pp. 389 -- 399

A CLASS OF ESTIMATORS FOR THE MEAN OF A FINITE POPULATION USING AUXILIARY INFORMATION

By

P. M. ROBINSON, London School of Economics

SUMMARY. A class of estimators for the mean of a finite population proposed by Srivastava (1971) is shown to nearly include a class of Srivastava and Jhajj (1981). Minimum-mean-squared-error members of the former's class are not computable if certain population parameters are unknown. With sample-based parameter estimators substituted, and certain modifications introduced, several such members are shown to achieve, to a first order of approximation, the mean-squared-error of the linear regression estimator that assumes knowledge of population regression coefficients, and the order of magnitude of the remainder term is derived, under assumptions on population parameters. Practical implications of the results are discussed.

AMS (1991) subject classification. 62D05.

Key words and phrases. Auxiliary information, simple random sampling, mean-squared-error, regression estimator.

Full paper (PDF)

This article in mathematical reviews.