An application of the Kiefer-Wolfowitz equivalence theorem to ANOVA models with additive regression (Q1116243)

For linear models of ANOVA type (discrete factors of influence) as well as for linear regression models (continuous factors of influence) an extensive theory of optimum designs has been developed. In this paper it is shown how results for these two types of models can be combined to find optimum designs with respect to various parameter subsets for more complex linear models with both discrete and continuous factors of influence. The criteria under consideraion are D- and \(D_ s\)- optimality.