DE- and EDPm- Compound Optimality for the Information and Probability-Based Criteria

Abstract

Several optimality criteria have been considered in the literature as information-based criteria. The probability- based criteria have been recently proposed for maximizing the probability of a desired outcome. However, designs that are optimal for the information- based criteria may be inadequate for probability- based criteria. This paper introduces the DE- and EDP${}_{M}$ -- optimum designs for multi aims of optimality for Generalized Linear Models (GLMs). An equivalence theorem is proved for both compound criteria. Finally, two numerical examples are given to illustrate the potentiality of the proposed compound criteria.