Sensitivity analysis in general metric spaces

被引：8

作者：

Gamboa, Fabrice ^{[1
,2
,3
]}

Klein, Thierry ^{[1
,4
,5
]}

Lagnoux, Agnes ^{[4
,6
,7
]}

Moreno, Leonardo ^{[8
]}

机构：

[1] Inst Math Toulouse, F-31062 Toulouse, France

[2] ANITI, F-31062 Toulouse, France

[3] Univ Toulouse, UMR5219, CNRS, UT3, F-31062 Toulouse, France

[4] Univ Toulouse, UMR5219, F-31058 Toulouse, France

[5] Univ Toulouse, ENAC Ecole Natl Aviat Civile, Toulouse, France

[6] Inst Math Toulouse, F-31058 Toulouse, France

[7] CNRS, UT2J, F-31058 Toulouse, France

[8] Univ Republica, FCEA, Dept Metodos Cuantitat, Montevideo, Uruguay

来源：

RELIABILITY ENGINEERING & SYSTEM SAFETY | 2021年 / 212卷

关键词：

Sensitivity analysis; Cramer-von-Mises distance; Pick-Freeze method; U-statistics; General metric spaces; MODEL; VARIANCE; PLUME;

D O I：

10.1016/j.ress.2021.107611

中图分类号：

T [工业技术];

学科分类号：

08 ;

摘要：

Sensitivity indices are commonly used to quantity the relative influence of any specific group of input variables on the output of a computer code. In this paper, we introduce new sensitivity indices adapted to outputs valued in general metric spaces. This new class of indices encompasses the classical ones; in particular, the so-called Sobol indices and the Cramer-von-Mises indices. Furthermore, we provide asymptotically Gaussian estimators of these indices based on U-statistics. Surprisingly, we prove the asymptotic normality straightforwardly. Finally, we illustrate this new procedure on a toy model and on two real-data examples.

引用

页数：8

共 37 条

[1] A new uncertainty importance measure [J].

Borgonovo, E. .

RELIABILITY ENGINEERING & SYSTEM SAFETY, 2007, 92 (06) :771-784

[2] Probabilistic sensitivity measures as information value [J].

Borgonovo, Emanuele ;

Hazen, Gordon B. ;

Jose, Victor Richmond R. ;

Plischke, Elmar .

EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2021, 289 (02) :595-610

[3] A Common Rationale for Global Sensitivity Measures and Their Estimation [J].

Borgonovo, Emanuele ;

Hazen, Gordon B. ;

Plischke, Elmar .

RISK ANALYSIS, 2016, 36 (10) :1871-1895

[4] Moment Independent Importance Measures: New Results and Analytical Test Cases [J].

Borgonovo, Emanuele ;

Castaings, William ;

Tarantola, Stefano .

RISK ANALYSIS, 2011, 31 (03) :404-428

[5] SENSITIVITY OF GAUSSIAN PLUME MODEL TO DISPERSION SPECIFICATIONS [J].

CARRASCAL, MD ;

PUIGCERVER, M ;

PUIG, P .

THEORETICAL AND APPLIED CLIMATOLOGY, 1993, 48 (2-3) :147-157

[6] Global sensitivity analysis with dependence measures [J].

Da Veiga, Sebastien .

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, 2015, 85 (07) :1283-1305

[7]

de Rocquigny E., 2008, Uncertainty in Industrial Practice: a Guide to Quantitative Uncertainty Management

[8] New sensitivity analysis subordinated to a contrast [J].

Fort, Jean-Claude ;

Klein, Thierry ;

Rachdi, Nabil .

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2016, 45 (15) :4349-4364

[9] SENSITIVITY INDICES FOR OUTPUT ON A RIEMANNIAN MANIFOLD [J].

Fraiman, R. ;

Gamboa, F. ;

Moreno, L. .

INTERNATIONAL JOURNAL FOR UNCERTAINTY QUANTIFICATION, 2020, 10 (04) :297-314

[10] Statistical inference for Sobol pick-freeze Monte Carlo method [J].

Gamboa, F. ;

Janon, A. ;

Klein, T. ;

Lagnoux, A. ;

Prieur, C. .

STATISTICS, 2016, 50 (04) :881-902

← 1 2 3 4 →