Premium
Sequential design for achieving estimated accuracy of global sensitivities
Author(s) -
Guenther John,
Lee Herbert K. H.,
Gray Genetha A.
Publication year - 2014
Publication title -
applied stochastic models in business and industry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.413
H-Index - 40
eISSN - 1526-4025
pISSN - 1524-1904
DOI - 10.1002/asmb.2091
Subject(s) - computer science , variable (mathematics) , sensitivity (control systems) , sampling (signal processing) , process (computing) , black box , design of experiments , gaussian process , function (biology) , gaussian , algorithm , mathematical optimization , statistics , mathematics , artificial intelligence , electronic engineering , engineering , mathematical analysis , physics , filter (signal processing) , quantum mechanics , evolutionary biology , biology , computer vision , operating system
Global sensitivity analysis provides information on the relative importance of the input variables for simulator functions used in computer experiments. It is more conclusive than screening methods for determining if a variable is influential, especially if a variable's influence is derived from its interactions with other variables. In this paper, we develop a method for providing global sensitivities with estimated accuracy. A treed Gaussian process serves as a statistical emulator of the black box function. A sequential experimental design makes effective and efficient use of simulator evaluations by adaptively sampling points that are expected to provide the maximum improvement to the emulator model. The method accounts for both sampling error and emulator error. Copyright © 2014 John Wiley & Sons, Ltd.