Non-parametric estimation of the first-order Sobol indices with bootstrap bandwidth. (English) Zbl 1497.62088
Summary: Suppose that \(Y=\psi(X_1,\dots,X_p)\), where \((X_1,\dots,X_p)^\top\) are random inputs, \(Y\) is the output, and \(\psi(\cdot)\) is an unknown link function. The Sobol indices gauge the sensitivity of each \(X\) against \(Y\) by estimating the regression curve’s variability between them. In this paper, we estimate these curves with a kernel-based method. The method allows to estimate the first order indices when the link between the independent and dependent variables is unknown. The kernel-based methods need a bandwidth to average the observations. For finite samples, the cross-validation method is famous to decide this bandwidth. However, it produces a structural bias. To remedy this, we propose a bootstrap procedure which reconstruct the model residuals and re-estimate the non-parametric regression curve. With the new set of curves, the procedure corrects the bias in the Sobol index. To test the developed method, we implemented simulated numerical examples with complex functions.
MSC:
| 62G07 | Density estimation |
| 62G08 | Nonparametric regression and quantile regression |
| 62G09 | Nonparametric statistical resampling methods |
Keywords:
Sobol indices; sensitivity analysis; nonparametric estimator; finite-sample bias; bootstrap bandwidthReferences:
| [1] | Alex Mara, T.; Rakoto Joseph, O., Comparison of some efficient methods to evaluate the main effect of computer model factors, Journal of Statistical Computation and Simulation, 78, 2, 167-78 (2008) · Zbl 1136.62416 · doi:10.1080/10629360600964454 |
| [2] | Baudin, M.; Boumhaout, K.; Delage, T.; Iooss, B.; Martinez, J.-M., Numerical stability of Sobol’ indices estimation formula, 50-51 (2016), Reunion Island, France |
| [3] | Box, G. E. P.; Draper, N. R., Wiley Series in probability and mathematical statistics: Applied probability and statistics, Empirical model-building and response surfaces (1987), John Wiley & Sons · Zbl 0614.62104 |
| [4] | Campolongo, F.; Saltelli, A.; Cariboni, J., From screening to quantitative sensitivity analysis. A unified approach, Computer Physics Communications, 182, 4, 978-88 (2011) · Zbl 1219.93120 · doi:10.1016/j.cpc.2010.12.039 |
| [5] | Carmichael, G. R.; Sandu, A.; Potra, F. A., Sensitivity analysis for atmospheric chemistry models via automatic differentiation, Atmospheric Environment, 31, 3, 475-89 (1997) · doi:10.1016/S1352-2310(96)00168-9 |
| [6] | Cukier, R.; Levine, H.; Shuler, K., Nonlinear sensitivity analysis of multiparameter model systems, Journal of Computational Physics, 26, 1, 1-42 (1978) · Zbl 0369.65023 · doi:10.1016/0021-9991(78)90097-9 |
| [7] | Cukier, R. I.; Fortuin, C. M.; Shuler, K. E.; Petschek, A. G.; Schaibly, J. H., Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients. I Theory, The Journal of Chemical Physics, 59, 8, 3873-8 (1973) · doi:10.1063/1.1680571 |
| [8] | Cullen, A. C.; Frey, H. C., Probabilistic techniques in exposure assessment (1999), Springer, US: Springer |
| [9] | de Rocquigny, É., La maîtrise des incertitudes dans un contexte industriel. 1re partie: une approche méthodologique globale basée sur des exemples, Journal de la société française de statistique, 147, 3, 33-71 (2006) · Zbl 1409.62256 |
| [10] | Draper, N. R.; Smith, H., Applied regression analysis, 2 (1981), Wiley · Zbl 0548.62046 |
| [11] | Faraway, J. J.; Jhun, M., Bootstrap choice of bandwidth for density estimation, Journal of the American Statistical Association, 85, 412, 1119-22 (1990) · doi:10.1080/01621459.1990.10474983 |
| [12] | Goos, P., Lecture notes in statistics, 164, The optimal design of blocked and split-plot experiments (2002), New York, NY: Springer, New York, NY · Zbl 1008.62068 |
| [13] | Hansen, B. E., Exact mean integrated squared error of higher order kernel estimators, Econometric Theory, 21, 6, 1031-57 (2005) · Zbl 1083.62032 · doi:10.1017/S0266466605050528 |
| [14] | Hardle, W.; Mammen, E., Comparing Nonparametric Versus Parametric Regression Fits, The Annals of Statistics, 21, 4, 1926-47 (1993) · Zbl 0795.62036 · doi:10.1214/aos/1176349403 |
| [15] | Härdle, W.; Werwatz, A.; Müller, M.; Sperlich, S., Nonparametric and semiparametric models. Springer Series in Statistics (2004), Berlin, Heidelberg: Springer Berlin Heidelberg, Berlin, Heidelberg · Zbl 1059.62032 |
| [16] | Hayfield, T.; Racine, J. S., Nonparametric econometrics: The np package, Journal of Statistical Software, 27, 5, 1-32 (2008) · doi:10.18637/jss.v027.i05 |
| [17] | Iooss, B.; Lemaître, P.; Dellino, G.; Meloni, C., Uncertainty management in simulation-optimization of complex systems: Algorithms and applications, A review on global sensitivity analysis methods, 101-122 (2015), Boston, MA: Springer US, Boston, MA · Zbl 1332.90003 |
| [18] | Ishigami, T.; Homma, T., An importance quantification technique in uncertainty analysis for computer models, 398-403 (1990), IEEE Comput. Soc. Press |
| [19] | Janon, A.; Klein, T.; Lagnoux, A.; Nodet, M.; Prieur, C., Asymptotic normality and efficiency of two Sobol index estimators, ESAIM: Probability and Statistics, 18, 3, 342-64 (2014) · Zbl 1305.62147 · doi:10.1051/ps/2013040 |
| [20] | Jansen, M. J., Analysis of variance designs for model output, Computer Physics Communications, 117, 1-2, 35-43 (1999) · Zbl 1015.68218 · doi:10.1016/S0010-4655(98)00154-4 |
| [21] | Loubes, J.-M.; Marteau, C.; Solís, M., Rates of convergence in conditional covariance matrix with nonparametric entries estimation, Communications in Statistics - Theory and Methods (2019) · Zbl 07528967 · doi:10.1080/03610926.2019.1602652 |
| [22] | Makowski, D.; Naud, C.; Jeuffroy, M.-H.; Barbottin, A.; Monod, H., Global sensitivity analysis for calculating the contribution of genetic parameters to the variance of crop model prediction, Reliability Engineering & System Safety, 91, 10-11, 1142-7 (2006) · doi:10.1016/j.ress.2005.11.015 |
| [23] | Myer, R.; Montgomery, D., Response surface methodology: Process and product optimization using designed experiments (2002), New York: Wiley, New York · Zbl 1161.62393 |
| [24] | Oakley, Jeremy E.; O’Hagan, Anthony, Probabilistic sensitivity analysis of complex models: A Bayesian approach, Journal of the Royal Statistical Society: Series B (Statistical Methodology), 66, 3, 751-69 (2004) · Zbl 1046.62027 · doi:10.1111/j.1467-9868.2004.05304.x |
| [25] | Owen, A. B., Better estimation of small sobol’ sensitivity indices, ACM Transactions on Modeling and Computer Simulation, 23, 2, 1-17 (2013) · Zbl 1490.62192 · doi:10.1145/2457459.2457460 |
| [26] | Padgett, W. J.; Thombs, L. A., Smooth nonparametric quantile estimation under censoring: Simulations and bootstrap methods, Communications in Statistics - Simulation and Computation, 15, 4, 1003-25 (1986) · Zbl 0609.62060 · doi:10.1080/03610918608812558 |
| [27] | Roxy, P.; Devore, J. L., Statistics: The exploration and analysis of data, 1-817 (2012), Pacific Grove, CA: COLE, Pacific Grove, CA |
| [28] | R Core Team, R: A language and environment for statistical computing (2018), Vienna, Austria: R Foundation for Statistical Computing, Vienna, Austria |
| [29] | Racine, J., Bias-corrected kernel regression, Journal of Quantitative Economics, 17, 1, 25-42 (2001) |
| [30] | Rall, L. B., Applications of software for automatic differentiation in numerical computation (1980), Springer, Vienna: Springer · Zbl 0439.68050 |
| [31] | Ratto, M.; Pagano, A., Using recursive algorithms for the efficient identification of smoothing spline ANOVA models, AStA Advances in Statistical Analysis, 94, 4, 367-88 (2010) · Zbl 1477.62098 · doi:10.1007/s10182-010-0148-8 |
| [32] | Romano, J. P., On weak convergence and optimality of kernel density estimates of the mode, The Annals of Statistics, 16, 2, 629-47 (1988) · Zbl 0658.62053 · doi:10.1214/aos/1176350824 |
| [33] | Saltelli, A., Making best use of model evaluations to compute sensitivity indices, Computer Physics Communications, 145, 2, 280-97 (2002) · Zbl 0998.65065 · doi:10.1016/S0010-4655(02)00280-1 |
| [34] | Saltelli, A.; Chan, K.; Scott, E. M., Sensitivity analysis, 134 (2000), New York: Wiley, New York · Zbl 0961.62091 |
| [35] | Saltelli, A.; Chan, K.; Scott, E. M., Sensitivity analysis (2009), New York: Wiley, New York |
| [36] | Saltelli, A.; Ratto, M.; Andres, T.; Campolongo, F.; Cariboni, J.; Gatelli, D.; Saisana, M.; Tarantola, S., Global sensitivity analysis. The primer (2007), Chichester, UK: John Wiley & Sons, Ltd, Chichester, UK |
| [37] | Sobol’, I., Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates, Mathematics and Computers in Simulation, 55, 1-3, 271-80 (2001) · Zbl 1005.65004 |
| [38] | Sobol’, I.; Tarantola, S.; Gatelli, D.; Kucherenko, S.; Mauntz, W., Estimating the approximation error when fixing unessential factors in global sensitivity analysis, Reliability Engineering & System Safety, 92, 7, 957-60 (2007) · doi:10.1016/j.ress.2006.07.001 |
| [39] | Sobol’, I. M., Sensitivity estimates for nonlinear mathematical models, Mathematical Modeling and Computational Experiment, 1, 4, 407-14 (1993) · Zbl 1039.65505 |
| [40] | Touati, T., Confidence intervals for Sobol’ indices, 93-4 (2016), Reunion Island, France |
| [41] | Tsybakov, A. B., Introduction to Nonparametric Estimation (2009), Springer Series in Statistics: Springer Series in Statistics, Springer New York, New York, NY · Zbl 1176.62032 |
| [42] | Zhao, Y.-Y.; Lin, J.-G.; Wang, H.-X., Robust bootstrap estimates in heteroscedastic semi-varying coefficient models and applications in analyzing Australia CPI data, Communications in Statistics - Simulation and Computation, 46, 4, 2638-53 (2017) · Zbl 1373.62266 · doi:10.1080/03610918.2015.1054940 |
| [43] | Zhu, L.-X. X.; Fang, K.-T. T., Asymptotics for kernel estimate of sliced inverse regression, The Annals of Statistics, 24, 3, 1053-68 (1996) · Zbl 0864.62027 · doi:10.1214/aos/1032526955 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
