Listy Biometryczne - Biometrical Letters Vol. 34(1997), No. 1, 27-34
The paper deals with three wide classes of optimality criteria of experimental designs: Universal Optimality, involving the classical definition of convexity of a function, Ge¬neral Optimality, involving a combination of differentiable functions and B-Universal Optimality, where the Schur convexity is used. Two theorems on relationships be¬tween pairs of these classes are proved and some corollaries drawn. The applicability of the considered criteria to designs under fixed and also mixed linear models makes the results general.
optimality criteria, experimental designs, information matrix