Admissible experimental designs in multiple polynomial regression (Q1194018): Difference between revisions

This paper provides connection between admissibility of a design in a multiple regression set up on \(\nu\) variables and admissibility of subdesigns in less than \(\nu\) variables. The admissibility is with respect to any design optimality criterion. It is shown that for a convex experimental region, an admissible linear regression design is concentrated on its extreme points, and an admissible quadratic regression design has at most one support point in the relative interior of an arbitrary face of the region.