Information matrices for non-maximal parameter subsystems for second-degree mixture experiments

Abstract

Many practical problems are associated with the investigation of mixture of m factors, assumed to influence the response through the proportions in which they are blended together. More often, the primary concern of the experimenter is to learn more about the subsystems of interest. This study adopts the second-degree Kronecker model put forward by Draper and Pukelsheim (1998) to derive the information matrices desired. The parameter subspace of interest in this study is nonmaximal parameter subsystem which is subspace of the full parameter space. This subspace is consistent with the estimability and feasibility conditions. Information matrices are obtained by employing the Kronecker model approach and applying the canonical unit vector definitions. Results for fro information matrices for m � 2 ingredients are obtained.