New improvements in the use of dependence measures for sensitivity analysis and screening

ABSTRACT Physical phenomena are commonly modelled by time consuming numerical simulators, function of many uncertain parameters whose influences can be measured via a global sensitivity analysis. The usual variance-based indices require too many simulations, especially as the inputs are numerous. To address this limitation, we consider recent advances in dependence measures, focusing on the distance correlation and the Hilbert–Schmidt independence criterion. We study and use these indices for a screening purpose. Numerical tests reveal differences between variance-based indices and dependence measures. Then, two approaches are proposed to use the latter for a screening purpose. The first approach uses independence tests, with existing asymptotic versions and spectral extensions; bootstrap versions are also proposed. The second considers a linear model with dependence measures, coupled to a bootstrap selection method or a Lasso penalization. Numerical experiments show their potential in the presence of many non-influential inputs and give successful results for a nuclear reliability application.

[1]  N. Aronszajn.  Theory of Reproducing Kernels. , 1950 .

[2]  Margaret J. Robertson,et al.  Design and Analysis of Experiments , 2006, Handbook of statistics.

[3]  C. Daniel.  131 Note: On Varying One Factor at a Time , 1958 .

[4]  Maurice G. Kendall,et al.  The Advanced Theory of Statistics, Vol. 2: Inference and Relationship , 1979 .

[5]  Robert E. Shannon,et al.  Design and analysis of simulation experiments , 1978, WSC '78.

[6]  D. Cacuci.  Sensitivity theory for nonlinear systems. I. Nonlinear functional analysis approach , 1981 .

[7]  Max D. Morris,et al.  Factorial sampling plans for preliminary computational experiments , 1991 .

[8]  Ilya M. Sobol,et al.  Sensitivity Estimates for Nonlinear Mathematical Models , 1993 .

[9]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[10]  A. Owen,et al.  Valuation of mortgage-backed securities using Brownian bridges to reduce effective dimension , 1997 .

[11]  Jack P. C. Kleijnen,et al.  Sensitivity analysis and related analyses: A review of some statistical techniques , 1997 .

[12]  Jack P. C. Kleijnen,et al.  Searching for important factors in simulation models with many factors: Sequential bifurcation , 1997 .

[13]  A. E. Hoerl,et al.  Ridge regression: biased estimation for nonorthogonal problems , 2000 .

[14]  H Christopher Frey,et al.  OF SENSITIVITY ANALYSIS , 2001 .

[15]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[16]  Bernhard Schölkopf,et al.  Measuring Statistical Dependence with Hilbert-Schmidt Norms , 2005, ALT.

[17]  Fuhui Long,et al.  Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy , 2003, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[19]  Max D. Morris,et al.  Sampling plans based on balanced incomplete block designs for evaluating the importance of computer model inputs , 2006 .

[20]  Jon C. Helton,et al.  Survey of sampling-based methods for uncertainty and sensitivity analysis , 2006, Reliab. Eng. Syst. Saf..

[21]  Le Song,et al.  A Kernel Statistical Test of Independence , 2007, NIPS.

[22]  Emanuele Borgonovo,et al.  A new uncertainty importance measure , 2007, Reliab. Eng. Syst. Saf..

[23]  Maria L. Rizzo,et al.  Measuring and testing dependence by correlation of distances , 2007, 0803.4101.

[24]  Stefano Tarantola,et al.  Uncertainty in Industrial Practice , 2008 .

[25]  Bruno Sudret,et al.  Global sensitivity analysis using polynomial chaos expansions , 2008, Reliab. Eng. Syst. Saf..

[26]  Saltelli Andrea,et al.  Global Sensitivity Analysis: The Primer , 2008 .

[27]  Bernhard Schölkopf,et al.  Kernel Choice and Classifiability for RKHS Embeddings of Probability Distributions , 2009, NIPS.

[28]  Hao Shen,et al.  Fast Kernel-Based Independent Component Analysis , 2009, IEEE Transactions on Signal Processing.

[29]  Olivier Roustant,et al.  Calculations of Sobol indices for the Gaussian process metamodel , 2008, Reliab. Eng. Syst. Saf..

[30]  Paola Annoni,et al.  Sixth International Conference on Sensitivity Analysis of Model Output How to avoid a perfunctory sensitivity analysis , 2010 .

[31]  Paola Annoni,et al.  Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index , 2010, Comput. Phys. Commun..

[32]  Sonja Kuhnt,et al.  Design and analysis of computer experiments , 2010 .

[33]  Bernhard Schölkopf,et al.  Kernel-based Conditional Independence Test and Application in Causal Discovery , 2011, UAI.

[34]  Kenji Fukumizu,et al.  Equivalence of distance-based and RKHS-based statistics in hypothesis testing , 2012, ArXiv.

[35]  Sébastien Da Veiga,et al.  Global sensitivity analysis with dependence measures , 2013, ArXiv.

[36]  Bernhard Schölkopf,et al.  The Randomized Dependence Coefficient , 2013, NIPS.

[37]  Krishnakumar Balasubramanian,et al.  Ultrahigh Dimensional Feature Screening via RKHS Embeddings , 2013, AISTATS.

[38]  Emanuele Borgonovo,et al.  Global sensitivity measures from given data , 2013, Eur. J. Oper. Res..

[39]  Wojciech Zaremba,et al.  Taxonomic Prediction with Tree-Structured Covariances , 2013, ECML/PKDD.

[40]  Matieyendou Lamboni,et al.  Derivative-based global sensitivity measures: General links with Sobol' indices and numerical tests , 2012, Math. Comput. Simul..

[41]  Alexandre Janon,et al.  Randomized pick-freeze for sparse Sobol indices estimation in high dimension , 2014, 1403.5537.

[42]  Elena Deza,et al.  Encyclopedia of Distances , 2014 .

[43]  Masashi Sugiyama,et al.  High-Dimensional Feature Selection by Feature-Wise Kernelized Lasso , 2012, Neural Computation.