Han Gang, Santner Thomas J, Rawlinson Jeremy J
H. Lee Moffitt Cancer Center & Research Institute, MRC/BIOSTAT, 12902 Magnolia Drive, Tampa, FL 33612, (
Technometrics. 2009 Nov 1;51(4):464-474. doi: 10.1198/TECH.2009.08126.
Tuning and calibration are processes for improving the representativeness of a computer simulation code to a physical phenomenon. This article introduces a statistical methodology for simultaneously determining tuning and calibration parameters in settings where data are available from a computer code and the associated physical experiment. Tuning parameters are set by minimizing a discrepancy measure while the distribution of the calibration parameters are determined based on a hierarchical Bayesian model. The proposed Bayesian model views the output as a realization of a Gaussian stochastic process with hyperpriors. Draws from the resulting posterior distribution are obtained by the Markov chain Monte Carlo simulation. Our methodology is compared with an alternative approach in examples and is illustrated in a biomechanical engineering application. Supplemental materials, including the software and a user manual, are available online and can be requested from the first author.
调优和校准是提高计算机模拟代码对物理现象代表性的过程。本文介绍了一种统计方法，用于在可从计算机代码和相关物理实验获得数据的情况下同时确定调优和校准参数。通过最小化差异度量来设置调优参数，而校准参数的分布则基于分层贝叶斯模型来确定。所提出的贝叶斯模型将输出视为具有超先验的高斯随机过程的一个实现。通过马尔可夫链蒙特卡罗模拟从所得后验分布中进行抽样。在示例中将我们的方法与另一种方法进行了比较，并在生物力学工程应用中进行了说明。补充材料，包括软件和用户手册，可在线获取，也可向第一作者索取。
