Dette, HolgerPepelyshev, Andrey2010-01-182010-01-182010-01-18http://hdl.handle.net/2003/2662110.17877/DE290R-8817The 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.enDiscussion Paper / SFB 823;01/2010Computer experimentImportant samplingJack-knifeRegressionSequential designUncertainty analysis310330620report