Let the design of an experiment be represented by an -dimensional vector of weights with nonnegative components. Let the quality of for the estimation of the parameters of the statistical model be measured by the criterion of -optimality, defined as the th root of the determinant of the information matrix , where are known matrices with rows. In this paper, we show that the criterion of -optimality is second-order cone representable. As a result, the method of second-order cone programming can be used to compute an approximate -optimal design with any system of linear constraints on the vector of weights. More importantly, the proposed characterization allows us to compute an exact -optimal design, which is possible thanks to high-quality branch-and-cut solvers specialized to solve mixed integer second-order cone programming problems. Our results extend to the case of the criterion of -optimality, which measures the quality of for the estimation of a linear parameter subsystem defined by a full-rank coefficient matrix . We prove that some other widely used criteria are also second-order cone representable, for instance, the criteria of -, -, - and -optimality. We present several numerical examples demonstrating the efficiency and general applicability of the proposed method. We show that in many cases the mixed integer second-order cone programming approach allows us to find a provably optimal exact design, while the standard heuristics systematically miss the optimum.
View on arXiv