Weighted A-optimal row-column designs for treatment-control comparisons

摘要

The problem of comparing v test treatments and a control treatment in a row-column design with b rows and k columns is considered. Two different sets of contrasts, test treatment-control and test treatment-test treatment contrasts, are considered. Conditions under which a design is weighted A-optimal for estimating these two sets of contrasts with unequal precision are derived. Actual construction of optimal designs will be studied elsewhere.