Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지

The Construction of an Efficient Incomplete Block Design by Almost Otrhogonal Latin Squares of Order 6

  • Dongwoo Kim (Department of Statistics, Sungkyunkwan University, 3-53, Myungryun-Dong, Chongro-Ku, Seoul, Korea, 110-745)
  • Published : 1997.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

The littice designs have prove efficient but they are not alwasy available. This article proposes an alternative, an almost lattice design, of the triple lattice design (v=36, k=6, r=4) which is not available. Here, we compare the almost lattice design to the .alpha.-design (v=36, k=6, r=4) which is another alternative of the triple lattice design (v=36, k=6, r=4). Consequently, we show the almost lattice design is a more efficient alternative than the $\alpha$-design through A-, D-, and E-optimality.

Keywords

References

  1. Optimum Experimental Designs Atkinson, A. C.;Donev, A. N.
  2. Technometric v.22 A Comparison of Algorithms for Constructing Exact D-Optimal Designs Cook, R. D.;Nachtsheim, C. J.
  3. Lecture Note on Experimental Designs Giesbrecht F. G.
  4. Journal of Combinatorial Theory v.A16 Sub-Latin squares and incomplete orthogonal arrays Horton J. D.
  5. Technometric v.20 An Algorithm for the Construction of D-Optimal Experimental Designs Mitchell T. J.
  6. Foundations of Optimum Experimental Design Pazman, Andrej
  7. Optimum Design of Experiments Pukelsheim, Friedrich
  8. SAS/QC Software(Version 6.10) The Optex Procedure Draft Manual SAS Institute Inc.
  9. Combinatorics of Experimental Design Street A. P.;Street, D. J.