Title: Efficient calibration for imperfect computer models

Journal Article · · Annals of Statistics
DOI: https://doi.org/10.1214/15-AOS1314 · OSTI ID:1405172
 [1];  [2]
  1. Chinese Academy of Sciences (CAS), Beijing (China); Georgia Tech
  2. Georgia Inst. of Technology, Atlanta, GA (United States)
+ Show Author Affiliations

Many computer models contain unknown parameters which need to be estimated using physical observations. Furthermore, the calibration method based on Gaussian process models may lead to unreasonable estimate for imperfect computer models. In this work, we extend their study to calibration problems with stochastic physical data. We propose a novel method, called the L2 calibration, and show its semiparametric efficiency. The conventional method of the ordinary least squares is also studied. Theoretical analysis shows that it is consistent but not efficient. Here, numerical examples show that the proposed method outperforms the existing ones.

Research Organization:
Georgia Tech Research Corp., Atlanta, GA (United States)
Sponsoring Organization:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
Grant/Contract Number:
SC0010548
OSTI ID:
1405172
Report Number(s):
DOE-GT-0010548-4; FG02-13ER26159
Journal Information:
Annals of Statistics, Journal Name: Annals of Statistics Journal Issue: 6 Vol. 43; ISSN 0090-5364
Publisher:
Institute of Mathematical StatisticsCopyright Statement
Country of Publication:
United States
Language:
English

References (24)

A Generalized Representer Theorem book January 2001
Introduction to Empirical Processes and Semiparametric Inference book January 2008
Reproducing Kernel Hilbert Spaces in Probability and Statistics book January 2004
The Design and Analysis of Computer Experiments book January 2003
Cases for the nugget in modeling computer experiments journal December 2010
Response Surface Methodology—Current Status and Future Directions journal January 1999
Statistical Adjustments to Engineering Models journal October 2009
Prediction and Computer Model Calibration Using Outputs From Multifidelity Simulators journal November 2013
Computer Model Calibration Using the Ensemble Kalman Filter journal November 2013
On the Choice of Nugget in Kriging Modeling for Deterministic Computer Experiments journal January 2014
Inverse problems as statistics journal May 2002
A methodology for probabilistic predictions of regional climate change from perturbed physics ensembles
  • Murphy, J. M.; Booth, B. B. B.; Collins, M.
  • Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 365, Issue 1857 https://doi.org/10.1098/rsta.2007.2077
journal June 2007
Bayesian calibration of computer models journal August 2001
Combining Field Data and Computer Simulations for Calibration and Prediction journal January 2004
Probabilistic Formulations for Transferring Inferences from Mathematical Models to Physical Systems journal January 2004
A Framework for Validation of Computer Models journal May 2007
Computer Model Calibration Using High-Dimensional Output journal June 2008
Bayesian Validation of Computer Models journal November 2009
Simultaneous Determination of Tuning and Calibration Parameters for Computer Experiments journal November 2009
Computer model validation with functional output journal October 2007
On methods of sieves and penalization journal December 1997
Penalized quasi-likelihood estimation in partial linear models journal June 1997
Optimal Global Rates of Convergence for Nonparametric Regression journal December 1982
Some Guidelines For Using Nonparametric Methods For Modeling Data From Response Surface Designs journal May 2005

Cited By (12)

Nonparametric estimation of probabilistic sensitivity measures journal August 2019
Opportunities and challenges of quality engineering for additive manufacturing journal July 2018
Model Calibration With Censored Data journal June 2017
Sequential Design for Functional Calibration of Computer Models journal June 2018
Inference for Errors-in-Variables Models in the Presence of Systematic Errors with an Application to a Satellite Remote Sensing Campaign journal July 2018
Bayesian Calibration of Inexact Computer Models journal July 2016
On Prediction Properties of Kriging: Uniform Error Bounds and Robustness journal May 2019
A Generalized Gaussian Process Model for Computer Experiments With Binary Time Series journal May 2019
Uncertainty Quantification in Case of Imperfect Models: A Non-Bayesian Approach: Uncertainty quantification journal January 2018
On Prediction Properties of Kriging: Uniform Error Bounds and Robustness preprint January 2017
Inference for Errors-in-Variables Models in the Presence of Systematic Errors with an Application to a Satellite Remote Sensing Campaign text January 2018
Inference for Errors-in-Variables Models in the Presence of Systematic Errors with an Application to a Satellite Remote Sensing Campaign text January 2018

Linked Research (1)

Similar Records

Model Calibration with Censored Data
Journal Article · 2017 · Technometrics · OSTI ID:1405186
+1 more
A Theoretical Framework for Calibration in Computer Models: Parametrization, Estimation and Convergence Properties
Journal Article · 2016 · SIAM/ASA Journal on Uncertainty Quantification · OSTI ID:1405142
A frequentist approach to computer model calibration
Journal Article · 2016 · Journal of the Royal Statistical Society: Series B (Statistical Methodology) · OSTI ID:1375159

Related Subjects

97 MATHEMATICS AND COMPUTING