Certified metamodels for sensitivity indices estimation

Janon, Alexandre, Nodet, Maëlle, and Prieur, Clémentine. Denis Poisson, Fédération, and Trélat, E., eds. "Certified metamodels for sensitivity indices estimation." ESAIM: Proceedings 35 (2012): 234-238. <http://eudml.org/doc/251271>.

@article{Janon2012,
abstract = {Global sensitivity analysis of a numerical code, more specifically estimation of Sobol indices associated with input variables, generally requires a large number of model runs. When those demand too much computation time, it is necessary to use a reduced model (metamodel) to perform sensitivity analysis, whose outputs are numerically close to the ones of the original model, while being much faster to run. In this case, estimated indices are subject to two kinds of errors: sampling error, caused by the computation of the integrals appearing in the definition of the Sobol indices by a Monte-Carlo method, and metamodel error, caused by the replacement of the original model by the metamodel. In cases where we have certified bounds for the metamodel error, we propose a method to quantify both types of error, and we compute confidence intervals for first-order Sobol indices.},
author = {Janon, Alexandre, Nodet, Maëlle, Prieur, Clémentine},
editor = {Denis Poisson, Fédération, Trélat, E.},
journal = {ESAIM: Proceedings},
language = {eng},
month = {4},
pages = {234-238},
publisher = {EDP Sciences},
title = {Certified metamodels for sensitivity indices estimation},
url = {http://eudml.org/doc/251271},
volume = {35},
year = {2012},
}

TY - JOUR
AU - Janon, Alexandre
AU - Nodet, Maëlle
AU - Prieur, Clémentine
AU - Denis Poisson, Fédération
AU - Trélat, E.
TI - Certified metamodels for sensitivity indices estimation
JO - ESAIM: Proceedings
DA - 2012/4//
PB - EDP Sciences
VL - 35
SP - 234
EP - 238
AB - Global sensitivity analysis of a numerical code, more specifically estimation of Sobol indices associated with input variables, generally requires a large number of model runs. When those demand too much computation time, it is necessary to use a reduced model (metamodel) to perform sensitivity analysis, whose outputs are numerically close to the ones of the original model, while being much faster to run. In this case, estimated indices are subject to two kinds of errors: sampling error, caused by the computation of the integrals appearing in the definition of the Sobol indices by a Monte-Carlo method, and metamodel error, caused by the replacement of the original model by the metamodel. In cases where we have certified bounds for the metamodel error, we propose a method to quantify both types of error, and we compute confidence intervals for first-order Sobol indices.
LA - eng
UR - http://eudml.org/doc/251271
ER -