E(s^2)-optimal and minimax-optimal cyclic supersaturated designs via multi-objective simulated annealing

Koukouvinos, Christos, Mylona, Kalliopi and Simos, Dimitris E. (2008) E(s^2)-optimal and minimax-optimal cyclic supersaturated designs via multi-objective simulated annealing. Journal of Statistical Planning and Inference, 138 (6), 1639-1646. (doi:10.1016/j.jspi.2007.05.044).

Record type: Article

Abstract

In this paper, we are interested in finding E(s2)E(s2)-optimal and minimax-optimal, two level cyclic structured supersaturated designs through a metaheuristic approach guided via multi-objective simulated annealing (SA). Our construction method is based on cyclic generators. This class of metaheuristics enabled us to build supersaturated designs for q=2,4,…,14q=2,4,…,14 generators. Comparisons are made with previous works and it is shown that SA gives promising results for supersaturated designs that satisfy more than one optimality property. Furthermore, we provide some lower bounds and explicit formulas for the frequency of the elements with maximum absolute values that appear in the information matrix, when these values are 2, 4 or 6.

This record has no associated files available for download.

More information

e-pub ahead of print date: 12 August 2007

Published date: 1 July 2008

Keywords: cyclic generator, E(s2)E(s2)-optimality, factorial designs, metaheuristics, minimax-optimality, multi-objective optimization, simulated annealing, supersaturated designs

Organisations: Statistics