Algorithmic construction of optimal designs on compact sets for concave and differentiable criteria

Abstract : We consider the problem of construction of optimal experimental designs (approximate theory) on a compact subset X of Rd with nonempty interior, for a concave and Lipschitz diff erentiable design criterion Phi (.) based on the information matrix. The proposed algorithm combines (a) convex optimization for the determination of optimal weights on a support set, (b) sequential updating of this support using local optimization, and (c) finding new support candidates using properties of the directional derivative of Phi(.). The algorithm makes use of the compactness of X and relies on a fi nite grid Xl C X for checking optimality. By exploiting the Lipschitz continuity of the directional derivatives of Phi(.), effi ciency bounds on X are obtained and epsilon-optimality on X is guaranteed. The eff ectiveness of the method is illustrated on a series of examples.
Liste complète des métadonnées

Cited literature [52 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01001706
Contributor : Luc Pronzato <>
Submitted on : Wednesday, June 4, 2014 - 5:09:03 PM
Last modification on : Monday, November 5, 2018 - 3:52:02 PM
Document(s) archivé(s) le : Thursday, September 4, 2014 - 1:05:35 PM

File

Opt_design-V5.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Luc Pronzato, Anatoly Zhigljavsky. Algorithmic construction of optimal designs on compact sets for concave and differentiable criteria. Journal of Statistical Planning and Inference, Elsevier, 2014, 154, pp.141-155. ⟨10.1016/j.jspi.2014.04.005⟩. ⟨hal-01001706⟩

Share

Metrics

Record views

362

Files downloads

193