The Minimax Admissibility Characterization Of Homogeneous Linear Estimates For Regression Coefficients Under Quadratic Loss Function

In this paper, we review the element theory of the linear model.The writer ingists: the minimax admissibility characterization ofhomogeneous linear estimates for regression coefficients underquadratic loss function.First, when V>O, the necessary and sufficient conditions aregiven for a linear admissible estimate and minimax admissibleestimate in the class of homogeneous estimates. Moreovef, weconclude the relation of the admissibility characterization and theminimax admissibility characterization. Finally, when VO, thenecessary condition is given for a linear minimax admissible estimatein the class of homogeneous estimates. These theories play animportant role in the reality.