The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
Activity Number:
|
538
|
Type:
|
Roundtables
|
Date/Time:
|
Wednesday, August 3, 2011 : 12:30 PM to 1:50 PM
|
Sponsor:
|
Section on Statistical Computing
|
Abstract - #302275 |
Title:
|
CANCELLED: Smoothing Splines with Product Kernels in Modeling of Computer Experiments
|
Author(s):
|
Zhanyang Zhang
|
Companies:
|
University of Michigan
|
Address:
|
|
Keywords:
|
Computer experiments ;
Gaussian Process ;
Smoothing Splines ;
Bayesian method ;
Kernel functions ;
Krigging
|
Abstract:
|
In many scientific studies, computer codes are used to simulate complicated physical processes. These computer codes are usually referred as computer experiments. One typical difficult of running computer experiments is the computational cost, in which one run can take days to finish. Thus there is a crucial need for efficient statistical surrogate models to predict the computational model using limited number of runs. One widely used statistical surrogate is the Bayesian Gaussian Process model, which utilizes the key assumption that the output of a computer experiment is a realization of Gaussian Process. In this work, we tried to relax the Gaussian Process assumption by putting the prediction problem into a regression framework. The proposed method here is called Smoothing Splines with product kernels. Smoothing spline models are flexible nonparametric regression models which are particularly suitable for modeling of computer experiments. In this work, we also compared the commonly used Bayesian Gaussian Process models with the proposed product smoothing spline models and showed that the proposed model is superior in terms of prediction error and model selection.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2011 program
|
2011 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.