Functional principal component analysis for global sensitivity analysis of model with spatial output

被引:23
|
作者
Perrin, T. V. E. [1 ]
Roustant, O. [2 ]
Rohmer, J. [3 ]
Alata, O. [4 ]
Naulin, J. P. [5 ]
Idier, D. [3 ]
Pedreros, R. [3 ]
Moncoulon, D. [5 ]
Tinard, P. [5 ]
机构
[1] Univ Clermont Auvergne, Inst Henri Fayol, CNRS, Mines St Etienne,UMR LIMOS 6158, F-42023 St Etienne, France
[2] Univ Toulouse, Inst Math Toulouse, INSA, 135 Ave Rangueil, F-31077 Toulouse 4, France
[3] Bur Rech Geol & Minieres, 3 Av Claude Guillemin,BP 36009, F-45060 Orleans 2, France
[4] Univ Lyon, Lab Hubert Curien, UMR CNRS 5516, UJM St Etienne,IOGS, F-42023 St Etienne, France
[5] CCR, 157 Blvd Haussmann, F-75008 Paris, France
关键词
Global sensitivity analysis; Spatial data; Functional principal component analysis; Wavelet; B-splines; SIMULATIONS; DEPENDENCE;
D O I
10.1016/j.ress.2021.107522
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Motivated by risk assessment of coastal flooding, we consider time-consuming simulators with a spatial output. The aim is to perform sensitivity analysis (SA), quantifying the influence of input parameters on the output. There are three main issues. First, due to computational time, standard SA techniques cannot be directly applied on the simulator. Second, the output is infinite dimensional, or at least high dimensional if the output is discretized. Third, the spatial output is non-stationary and exhibits strong local variations. We show that all these issues can be addressed all together by using functional PCA (FPCA). We first specify a functional basis, such as wavelets or B-splines, designed to handle local variations. Secondly, we select the most influential basis terms, either with an energy criterion after basis orthonormalization, or directly on the original basis with a penalized regression approach. Then FPCA further reduces dimension by doing PCA on the most influential basis coefficients, with an ad-hoc metric. Finally, fast-to-evaluate metamodels are built on the few selected principal components. They provide a proxy on which SA can be done. As a by-product, we obtain analytical formulas for variance-based sensitivity indices, generalizing known formula assuming orthonormality of basis functions.
引用
收藏
页数:13
相关论文
共 50 条
  • [1] Sensitivity analysis in functional principal component analysis
    Yamanishi, Y
    Tanaka, Y
    COMPUTATIONAL STATISTICS, 2005, 20 (02) : 311 - 326
  • [2] Sensitivity analysis in functional principal component analysis
    Yoshihiro Yamanishi
    Yutaka Tanaka
    Computational Statistics, 2005, 20 : 311 - 326
  • [3] Spatial Autocorrelation of Global Stock Exchanges Using Functional Areal Spatial Principal Component Analysis
    Khoo, Tzung Hsuen
    Pathmanathan, Dharini
    Dabo-Niang, Sophie
    MATHEMATICS, 2023, 11 (03)
  • [4] Principal component analysis of multivariate spatial functional data
    Si-ahmed, Idris
    Hamdad, Leila
    Agonkoui, Christelle Judith
    Kande, Yoba
    Dabo-Niang, Sophie
    BIG DATA RESEARCH, 2025, 39
  • [5] Implementation of supervised principal component analysis for global sensitivity analysis of models with correlated inputs
    Mohammad Ali Mohammad Jafar Sharbaf
    Mohammad Javad Abedini
    Stochastic Environmental Research and Risk Assessment, 2022, 36 : 2789 - 2818
  • [6] Implementation of supervised principal component analysis for global sensitivity analysis of models with correlated inputs
    Sharbaf, Mohammad Ali Mohammad Jafar
    Abedini, Mohammad Javad
    STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT, 2022, 36 (09) : 2789 - 2818
  • [7] Spatial functional principal component analysis with applications to brain image data
    Li, Yingxing
    Huang, Chen
    Haerdle, Wolfgang K.
    JOURNAL OF MULTIVARIATE ANALYSIS, 2019, 170 : 263 - 274
  • [8] Global Modular Principal Component Analysis
    Kadappa, Vijayakumar
    Negi, Atul
    SIGNAL PROCESSING, 2014, 105 : 381 - 388
  • [9] Robust functional principal component analysis via a functional pairwise spatial sign operator
    Wang, Guangxing
    Liu, Sisheng
    Han, Fang
    Di, Chong-Zhi
    BIOMETRICS, 2023, 79 (02) : 1239 - 1253
  • [10] A survey of functional principal component analysis
    Shang, Han Lin
    ASTA-ADVANCES IN STATISTICAL ANALYSIS, 2014, 98 (02) : 121 - 142