Identifying and Cataloging Minimal SOS Designs
Identifying and Cataloging Minimal SOS Designs Senior Supervisor: Boxin Tang
Second order saturated (SOS) designs allow all their degrees of freedom to be used in estimating main effects and two-factor interactions. Obviously, adding a factor to an SOS design still gives an SOS design. An SOS design is said to be minimal if it is no longer minimal after any of its factors is deleted. Thus, all SOS designs can be obtained if we know all minimal SOS designs. The main objective of this project is to explore, identify and catalog all SOS designs for small run sizes. The secondary objective is to find some good SOS designs under the minimum aberration or other robust design criteria. Having STAT 430 is a plus but not required. The project will involve some paper reading and R programming.