Optimal designs for discriminating dose response models in toxicology studies

Abstract

We consider design issues for toxicology studies when we have a continuous response but the true mean response is only known to be a member in a class of nested models. This class of models were proposed by toxicologists who were concerned with only estimation problems. We develop robust and effcient designs for model discrimination and optimal designs for estimating parameters in the selected model at the same time. In particular, we propose designs that maximize the minimum of D- or D_1-efficiencies over all models in the given class. We show that these optimal designs are efficient for determining an appropriate model from the postulated class, quite efficient for estimating model parameters in the identified model and also robust with respect to model mis-specification. To facilitate use of these designs in practice, we have also constructed a web site to enable practitioners to generate optimal designs for their problems.