Estimation of global sensitivity indices for models with dependent variables
Estimation of global sensitivity indices for models with dependent variables
复制标题
DOI：
10.1016/j.cpc.2011.12.020
复制
发表时间：
2012-04-01
影响因子：
6.3
通讯作者：
Annoni, P.
中科院分区：
文献类型：
--
作者：
Kucherenko, S.;Tarantola, S.;Annoni, P.
提出了一种新的估计含因变量模型的基于方差的灵敏度指标的方法。一阶灵敏度指数和总灵敏度指数都是Sobol灵敏度指数的推广。导出了与Sobol公式类似的公式和MonteCarlo数值估计。提出了一种基于Copula的任意多变量概率分布抽样方法。一个很好的协议之间的分析和数值的一阶和总指数所考虑的测试用例。敏感性指数的行为取决于相互作用和相关性的相对优势。该方法被证明是有效的和通用的。C）2011 Elsevier B.V.保留所有权利。
A novel approach for estimation variance-based sensitivity indices for models with dependent variables is presented. Both the first order and total sensitivity indices are derived as generalizations of Sobol' sensitivity indices. Formulas and Monte Carlo numerical estimates similar to Sobol' formulas are derived. A copula-based approach is proposed for sampling from arbitrary multivariate probability distributions. A good agreement between analytical and numerical values of the first order and total indices for considered test cases is obtained. The behavior of sensitivity indices depends on the relative predominance of interactions and correlations. The method is shown to be efficient and general. C) 2011 Elsevier B.V. All rights reserved.