Department of Statistics, The Ohio State University
Statistics and Biostatistics Colloquium Series

Supersaturated designs are used for experiments in which a few important factors (variables) need to be identified out of a large number of potentially important factors, but only relatively small number of observations can be taken.

A class of supersaturated designs called k-circulant designs will be described. These designs are constructed from cyclic generators by cycling k elements at a time. The class of designs includes many optimal designs, some of which are already known and some of which are more efficient than known designs. Projection properties of some of the designs (for fitting models for a few factor effects) are explored.

The talk also illustrates that some k-circulant supersaturated designs can be augmented with interaction columns to produce efficient designs for a larger number of factors or for estimating interactions.