Stochastic kriging

Stochastic kriging is a metamodeling methodology developed for stochastic simulation experiments; it is based on the highly successful kriging method for the design and analysis of computer experiments. Stochastic kriging distinguishes the uncertainty about the performance response surface from the sampling uncertainty inherent in the stochastic simulation; it accomplishes this by incorporating trend models that are common in least-squares regression, spatial correlation modeling to account for inadequacies of the trend model, and an output-variance model which is also based on spatial correlation concepts. Stochastic kriging facilitates adaptive, sequential experiment designs that systematically reduce both model and sampling uncertainty to a user-specified level. The stochastic kriging Matlab software available from this web site is distributed ”as is,” without warranties of any kind, either express or implied. Altough no user support is provide, please report problems via the contact link.

Keywords for this software

Anything in here will be replaced on browsers that support the canvas element


References in zbMATH (referenced in 1 article )

Showing result 1 of 1.
Sorted by year (citations)

  1. Couckuyt, Ivo; Dhaene, Tom; Demeester, Piet: ooDACE toolbox: a flexible object-oriented Kriging implementation (2014)