In this paper the problem of estimating regression coefficients in seemingly unrelated regression equations under some quadratic losses and matrix losses is discussed. The necessary and sufficient conditions for the existence of the uniformly minimum risk equivariant estimator under an affine group and a transitive group of transformations are given respectively. The minimaxity of the least squares estimators are also studied.
|Number of pages||7|
|Journal||Statistics & Decisions|
|Publication status||Published - 1 Feb 1997|