Robust Improvement in Estimation of a Mean Matrix in an Elliptically Contoured Distribution

In estimation of a matrix of regression coefficients in a multivariate linear regression model, this paper shows that minimax and shrinkage estimators under a normal distribution remain robust under an elliptically contoured distribution. The robustness of the improvement is established for both invariant and noninvariant loss functions in the above model as well as in the growth curve model.