A computational algebraic approach for saturated D-optimal designs with n ≡ 2 (mod 4) observations

Abstract

The use of D-optimal designs in experimental situations has the advantage of minimizing the volume of the confidence ellipsoid for the estimates β̂ of a first order linear model. We apply Computational Algebra methods to the construction of saturated D-optimal designs for n ≡ 2 (mod 4) observations.