A branch-and-bound algorithm for the exact optimal experimental design problem

We discuss a generalisation of the approximate optimal experimental design problem, in which weight each regression point needs to stay closed interval. work with Kiefer’s optimality criteria include well-known D- and A-optimality as special cases. propose first-order algorithm for generalised problem that redistributes weights two points iteration. develop branch-and-bound exact problems under where subproblems search tree are equivalent generalized therefore, can be solved efficiently by method. observe our is favourable popular exchange heuristic certain instances.