Extending Morris Method: identification of the interaction graph using cycle-equitabe designs - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Statistical Computation and Simulation Année : 2015

Extending Morris Method: identification of the interaction graph using cycle-equitabe designs

Résumé

The paper presents designs that allow detection of mixed effects when performing preliminary screening of the inputs of a scalar function of $d$ input factors, in the spirit of Morris' Elementary Effects approach. We introduce the class of $(d,c)$-cycle equitable designs as those that enable computation of exactly $c$ second order effects on all possible pairs of input factors. Using these designs, we propose a fast Mixed Effects screening method, that enables efficient identification of the interaction graph of the input variables. Design definition is formally supported on the establishment of an isometry between sub-graphs of the unit cube $Q_d$ equipped of the Manhattan metric, and a set of polynomials in $(X_1,\ldots, X_d)$ on which a convenient inner product is defined. In the paper we present systems of equations that recursively define these $(d,c)$-cycle equitable designs for generic values of $c\geq 1$, from which direct algorithmic implementations are derived. Application cases are presented, illustrating the application of the proposed designs to the estimation of the interaction graph of specific functions.
Fichier principal
Vignette du fichier
SCSpaperFedou.pdf (4.1 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01318096 , version 1 (19-05-2016)

Identifiants

Citer

Jean-Marc Fédou, Maria João Torres Dolores Rendas. Extending Morris Method: identification of the interaction graph using cycle-equitabe designs. Journal of Statistical Computation and Simulation, 2015, 85 (7), pp. 1281-1282. ⟨10.1080/00949655.2015.1008226⟩. ⟨hal-01318096⟩
175 Consultations
247 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More