A Characterization of Admissible Linear Estimator of Regression Coefficients in Variance Component Models

In the paper, for the variance component models we take the ordinary quadratic risk function, and consider the admissibility of the linear estimators of linear combinations of regression coefficients in the class of linear homogeneous and inhomogeneous estimators. We get the necessary and sufficient conditions for the linear estimators of linear combinations of regression coefficients to be admissible.