A Comparison Study of Different Optimizing Criteria and Confounding Patterns For Multi-Level Binary Replacement and Other Designs Used in Computer Experiments
We have constructed four different types of designs for computer experiments. The
design types are based on latin hypercube sampling (LHS), orthogonal arrays (OA), ran-
dom sampling and the recently proposed multi-level binary replacement (MBR) design.
For each type of design we have attempted to find the best possible design out of a
certain number of constructed designs using three different optimizing criteria: the alias sum of square criterion (ASSC), the L-criterion and a modified A-criterion. The chosen design has then been tested by fitting an approximate model and calculating maximum error (MAX) and root mean squared error (RMSE) values. We observed that out of the three criteria applied the ASSC performed the best.
In addition to comparing criteria for optimizing the design choice, we have also
constructed non-optimized designs for comparing the different design types and the
different ways of constructing MBR designs. In this setting we observed that OA designs
performed well in general, whereas the MBR designs performed well when restricted to
a small number of factors.
Place, publisher, year, edition, pages
Institutt for matematiske fag , 2011. , 65 p.
ntnudaim:6349, MTFYMA fysikk og matematikk, Industriell matematikk
IdentifiersURN: urn:nbn:no:ntnu:diva-14418Local ID: ntnudaim:6349OAI: oai:DiVA.org:ntnu-14418DiVA: diva2:453360
Tyssedal, John Sølve, FørsteamanuensisMartens, Harald