Quantile and probability-level criteria for nonlinear experimental design

Andrej Pazman (1) , Luc Pronzato (2)

1 Dep. Applied Mathematics and Statistics, Faculty of Mathematics, Physics and Informatics
2 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe SYSTEMES

We consider optimal experimental design for parameter estimation in nonlinear situations where the optimal experiment depends on the value of the parameters to be estimated. Setting a prior distribution for these parameters, we construct criteria based on quantiles and probability levels of classical design criteria and show how their derivatives can easily be approximated, so that classical algorithms for local optimal design can be used for their optimisation.

Mots clés: robust design; minimax optimal design; average optimal design; quantiles; nonlinear models

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Andrej Pazman, Luc Pronzato. Quantile and probability-level criteria for nonlinear experimental design. 8th Int. Workshop in Model-Oriented Design and Analysis, Jun 2007, Almagro, Spain. pp.157-164. ⟨hal-00170634⟩