Exploration of D-, A-, I- and G- Optimality Criteria in Mixture Modeling
Samson W. Wanyonyi *
Department of Mathematics and Computer Science, University of Eldoret, Eldoret, Kenya.
Ayubu A. Okango
Department of Mathematics and Actuarial Science, Murang’a University of Technology, Murang’a, Kenya.
Julius K. Koech
Department of Mathematics and Computer Science, University of Eldoret, Eldoret, Kenya.
*Author to whom correspondence should be addressed.
Abstract
A design optimality criterion, such as D-, A-, I-, and G- optimality criteria, is often used to analyze, evaluate and compare different designs options in mixture modeling test. A mixture test is an experiment where the descriptive variable and response rely only on the mixture's relative ratio in the mix but not its composition. The study geared toward exploring D-, A-, I-, and G- optimality criteria and their efficiency in determining an optimal split-plot design in mixture modeling within the presences of process variables. We evaluated and discussed in detail D-, A-, I-, and G- optimality criteria based on literature review. We also explored and examine why I- and D-optimal criteria are often involved within the formulation of an optimal design in the context of mixture process variable settings. We recommend that optimality criterion must always be used when assessing the various styles of designs so as to search out a desirable design that matches a combination model.
Keywords: mixture designs, optimality criteria, optimal designs, split-plot designs, process variable