**Sankhya: The Indian Journal of Statistics**

1995, Volume 57, Series B, Pt. 1, pp. 158--165

A NOTE ON THE STEIN RULE ESTIMATION IN LINEAR MODELS WITH NONSCALAR ERROR COVARIANCE MATRIX

By

ANOOP CHATURVEDI, *University of Allahabad*

SUMMARY. For the coefficient vector of a linear regression model with non-scalar error covariance matrix. Feasible Generalised Least Squares (FGLS) and Stein rule estimators are considered and taking sample size to be large, condition for the dominance of the Stein rule estimator over the FGLS estimator is obtained under a generalized Pitman nearness criterion.

AMS (1991) subject classification. 62J05, 62J07.

Key words and phrases. Feasible generalized least squares estimator(FGLS), Stein rule estimator, generalized Pitman nearness criterion, non-spherical disturbances, AR(1) disturbances.