Abstract
Recently, uncertainty quantification is getting more and more attention, especially for computer model calibration. However, most of the existing papers assume the errors follow a Gaussian or sub-Gaussian distribution, which would not be satisfied in practice. To overcome the limitation of the traditional calibration procedures, the authors develop a robust calibration procedure based on Huber loss, which can deal with responses with outliers and heavy-tail errors efficiently. The authors propose two different estimators of the calibration parameters based on ordinary least estimator and L2 calibration respectively, and investigate the nonasymptotic and asymptotic properties of the proposed estimators under certain conditions. Some numerical simulations and a real example are conducted, which verifies good performance of the proposed calibration procedure.
Similar content being viewed by others
References
Tuo R and Wu C F J, Efficient calibration for imperfect computer models, Annals of Statistics, 2015, 43(6): 2331–2352.
Wong K W, Storlie C B, and Lee C M, A frequentist approach to computer model calibration, Journal of Royal Statistical Society Series B, 2017, 79(2): 635–648.
Tuo R, Adjustments to computer models via projected kernel calibration, SIAM/ASA Journal on Uncertainty Quantification, 2019, 7(2): 553–578.
Xie F Z and Xu Y X, Bayesian projected calibration of computer models, Journal of American Statistical Association, 2021, 116(536): 1965–1982.
Wang Y and Tuo R, Semi-parametric adjustment to computer models, Statistics, 2020, 54(6): 1255–1275.
Kennedy M C and O’Hagan A, Bayesian calibration of computer models, Journal of Royal Statistical Society Series B, 2001, 63: 425–464.
Tuo R, Wang Y, and Wu C F J, On the improved rates of convergence for matern-type kernel ridge regression with application to calibration of computer models, SIAM/ASA Journal on Uncertainty Quantification, 2020, 8(4): 1522–1547.
Tuo R, He S Y, Pourhabib A, et al., A reproducing kernel hilbert space approach to functional calibration of computer models, Journal of American Statistical Association, 2021, DOI: https://doi.org/10.1080/01621459.2021.1956938.
Zhou W X, Bose K, Fan J Q, et al., A new perspective on robust M-estimations: Finite sample theory and applications to dependence-adjusted multiple testing, Annals of Statistics, 2018, 46(5): 1904–1931.
Huber P J, Robust estimation of a location parameter, Annals of Mathematical Statistics, 1964, 35: 73–101.
Santner T J, Williams B J, and Notz W I, The Design and Analysis of Computer Experiments, Springer, New York, 2003.
Gu C, Smoothing spline anova models: R package gss, Journal of Statictical Software, 2014, 58: 1–25.
van de Geer S, Empirical Processes in M-Estimation, Cambridge University Press, Cambridge, 2020.
Sun Q, Zhou W X, and Fan J Q, Adaptive huber regression, Journal of the American Statistical Association, 2020, 115(529): 254–265.
Spokoiny V, Parametric estimation, finite sample theory, Annals of Statistics, 2012, 40: 2877–2909.
Vershynin R, Introduction to the Non-Asymptotic Analysis of Random Matrices, Compressed Sensing 210–268, Cambridge University Press, Cambridge, 2012.
Spokoiny V, Bernstein-von Mises theorem for growing parameter dimension, 2013, Available at arXiv: 1302.3430.
Mammen E and Van de Geer S, Penalized quasi-likelihood estimation in partial linear models, Annals of Statistics, 1997, 25: 1014–1035.
Serfling R J, Approximation Theorems of Mathematical Statistics, John Wiley and Sons, New York, 1980.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The authors declare no conflict of interest.
Additional information
This research was supported by the Science Challenge Project under Grant No. TZ2018001.
Rights and permissions
About this article
Cite this article
Sun, Y., Fang, X. Robust Calibration of Computer Models Based on Huber Loss. J Syst Sci Complex 36, 1717–1737 (2023). https://doi.org/10.1007/s11424-023-1456-x
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11424-023-1456-x