NPUA: A new approach for the analysis of computer experiments
Loading...
Date
2010-01-18T10:55:05Z
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The main issue in the analysis of computer experiments is an uncertainty of prediction and related inferences. To address the uncertainty analysis, the Bayesian analysis of deterministic computer models has been actively developed in the last decade. In the Bayesian approach, the uncertainty is expressed through a Gaussian process model. As a consequence, the resulting analysis
is rather sensitive with respect to these prior assumptions. Moreover, for high
dimensional data this approach leads to time consuming computations. In the present paper we introduce a new approach for deriving the uncertainty in the analysis of computer experiments, where the distribution of uncertainty is obtained in a general nonparametric form. The proposed approach is called N(on) P(arametric) U(ncertainty) A(nalysis) and is based on a combination of
sampling and regression techniques. In particular, it is computationally very
simple. We compare NPUA with the Bayesian and Kriging method and investigate
its performance for finding points for the next runs by re-analyzing the ASET model.
Description
Table of contents
Keywords
Computer experiment, Important sampling, Jack-knife, Regression, Sequential design, Uncertainty analysis