Advertisement

Journal of Statistical Theory and Practice

, Volume 7, Issue 4, pp 725–731 | Cite as

A Note on Saturated First-Order Design with Foldover Structure

  • Dennis K. J. Lin
  • Lili Xiao
Article

Abstract

A first-order saturated design is the smallest design resulting in unbiased estimates for all main effects. It could be misleading in the presence of interaction effects. This article provides a simple class of first-order saturated designs in which specific two-factor interactions are orthogonal to many of the main effects, while keeping rather high design efficiencies.

Keywords

D-optimal design Foldover Interaction effects Main effect Saturated design 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Koukouvinos, C., M. Mitrouli, and J. Seberry. 2000. Bounds on the maximum determinant for (1, −1) matrices. Bull. Inst. Combin. Appl., 29, 39–48.MathSciNetzbMATHGoogle Scholar
  2. Li, W., and D. K. Lin. 2003. Optimal foldover two-level fractional factorial designs. Technometrics, 45, 142–149.MathSciNetCrossRefGoogle Scholar
  3. Lin, D. K. 1993. Another look at first-order saturated designs: The p-efficient designs. Technometrics, 35, 284–292.MathSciNetzbMATHGoogle Scholar
  4. Miller, A., and R. Sitter. 2005. Using folded over non-orthogonal designs. Technometrics, 47, 502–513.MathSciNetCrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2013

Authors and Affiliations

  1. 1.Department of StatisticsThe Pennsylvania State UniversityUniversity ParkUSA
  2. 2.Tsinghua UniversityBeijingChina

Personalised recommendations