Computing D-optimal Size-and-Cost Constrained Designs of Experiments
Sprache des Vortragstitels:
PROBASTAT 2015, The Seventh International Conference on Probability and Statistics
Sprache des Tagungstitel:
In this paper, we study the problem of D-optimal experimental design under two linear constraints, which can be interpreted as simultaneous restrictions on the size and on the total cost of the experiment. For computing a size-and-cost constrained approximate D-optimal design, we propose a ?barycentric? algorithm with sequential removal of redundant design points. We analytically prove convergence results about the barycentric algorithm and numerically demonstrate its favourable properties compared to competing methods.