Adjustments to Computer Models via Projected Kernel Calibration
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
Copyright by SIAM and ASA. Identification of model parameters in computer simulations is an important topic in computer ex- periments. We propose a new method, called the projected kernel calibration method, to estimate these model parameters. The proposed method is proven to be asymptotic normal and semipara- metric efficient. As a frequentist method, the proposed method is as efficient as the L2 calibration method proposed by Tuo and Wu [Ann. Statist., 43 (2015), pp. 2331{2352]. On the other hand, the proposed method has a natural Bayesian version, which the L2 method does not have. This Bayesian version allows users to calculate the credible region of the calibration parameters without using a large sample approximation. We also show that the inconsistency problem of the calibration method proposed by Kennedy and O'Hagan [J. R. Stat. Soc. Ser. B. Stat. Methodol., 63 (2001), pp. 425{464] can be rectified by a simple modification of the kernel matrix.