Minimax Second-Order Disigns over Hypercubes for the Difference Between Estimated Response at two Points on a Straight Line Through the Origin

Abstract

Minimization of the variance of the difference between estimated responses at two points on a straight line through the origin maximized over all pairs of such points in the experimental region is taken as the design criterion. Optimal designs are obtained for second-order polynomial models over hypercubes. The performance of best designs within some commonly used classes of designs is also investigated.