405 – Advances in Design and Analysis of Clinical Trials
Minimal Plus One Design Points to Test Lack of Fit for Second-Degree Mixture Model
Yanyan Li
Teva Branded Pharmaceutical Products R&D, Inc.
Damaraju Raghavarao
Temple University
D-optimal minimal designs have been obtained for different mixture models with minimal support design points, i.e., the number of design points is equal to the number of parameters in the model. To test Lack of Fit, we need to add at least one additional distinct design point. Those distinct design points are within the design space with all factor levels greater than zero for practical reasons. We will discuss second-degree mixture model and second-degree mixture model with main effects and two factor interactions including one common factor. The new design will be compared with other possible designs for testing Lack of Fit.