Uniform semi-Latin squares and their pairwise-variance aberrations

DOI10.1016/J.JSPI.2020.12.003MaRDI QIDQ830735zbMATH OpenOpenAlexFDO

Authors R. A. Bailey, Leonard H. Soicher

Publication date 7 May 2021

Published in Journal of Statistical Planning and Inference (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2008.10552

zbMATH Keywords

block designorthogonal arraySchur optimalityaffine resolvable designbalanced incomplete-block designdesign optimality

Mathematics Subject Classification ID

Optimal statistical designs (62K05) Statistical block designs (62K10) Combinatorial aspects of block designs (05B05) Orthogonal arrays, Latin squares, Room squares (05B15)

Abstract: For integers $n>2$ and $k>0$, an $(nimesn)/k$ semi-Latin square is an $nimesn$ array of $k$-subsets (called blocks) of an $nk$-set (of treatments), such that each treatment occurs once in each row and once in each column of the array. A semi-Latin square is uniform if every pair of blocks, not in the same row or column, intersect in the same positive number of treatments. We show that when a uniform $(nimesn)/k$ semi-Latin square exists, the Schur optimal $(nimesn)/k$ semi-Latin squares are precisely the uniform ones. We then compare uniform semi-Latin squares using the criterion of pairwise-variance (PV) aberration, introduced by J.P. Morgan for affine resolvable designs, and determine the uniform $(nimesn)/k$ semi-Latin squares with minimum PV aberration when there exist $n-1$ mutually orthogonal Latin squares (MOLS) of order $n$. These do not exist when $n=6$, and the smallest uniform semi-Latin squares in this case have size $(6imes6)/10$. We present a complete classification of the uniform $(6imes6)/10$ semi-Latin squares, and display the one with least PV aberration. We give a construction producing a uniform $((n+1)imes(n+1))/((n-2)n)$ semi-Latin square when there exist $n-1$ MOLS of order $n$, and determine the PV aberration of such a uniform semi-Latin square. Finally, we describe how certain affine resolvable designs and balanced incomplete-block designs (BIBDs) can be constructed from uniform semi-Latin squares. From the uniform $(6imes6)/10$ semi-Latin squares we classified, we obtain (up to block design isomorphism) exactly 16875 affine resolvable designs for 72 treatments in 36 blocks of size 12 and 8615 BIBDs for 36 treatments in 84 blocks of size 6. In particular, this shows that there are at least 16875 pairwise non-isomorphic orthogonal arrays $mathrmOA(72,6,6,2)$.

Recommendations

• Uniform semi-Latin squares and their Schur-optimality
• On uniformly generating Latin squares
• scientific article; zbMATH DE number 2104809
• Large deviations in random latin squares
• On the existence of latin squares with special distribution properties
• Intercalates and discrepancy in random Latin squares
• A note on semi-Latin squares
• scientific article; zbMATH DE number 3967393
• Marginal homogeneity and the superposition of Latin squares

Cites work

• scientific article; zbMATH DE number 4015963 (Why is no real title available?)
• scientific article; zbMATH DE number 5711161 (Why is no real title available?)
• scientific article; zbMATH DE number 3951837 (Why is no real title available?)
• scientific article; zbMATH DE number 3550719 (Why is no real title available?)
• scientific article; zbMATH DE number 3637149 (Why is no real title available?)
• scientific article; zbMATH DE number 775912 (Why is no real title available?)
• scientific article; zbMATH DE number 798642 (Why is no real title available?)
• scientific article; zbMATH DE number 847275 (Why is no real title available?)
• scientific article; zbMATH DE number 3048068 (Why is no real title available?)
• scientific article; zbMATH DE number 3097793 (Why is no real title available?)
• 2-\((v,k,\lambda)\) designs of small order
• A family of balanced incomplete-block designs with repeated blocks on which general linear groups act
• Algorithms for finding generalized minimum aberration designs
• Classification of orthogonal arrays by integer programming
• Combinatorics of optimal designs
• Complete enumeration of pure-level and mixed-level orthogonal arrays
• Construction and optimality of affine-resolvable designs
• Designs, groups and computing
• Engineering an efficient canonical labeling tool for large and sparse graphs
• Enumeration of MOLS of small order
• Finding the symmetry group of an LP with equality constraints and its application to classifying orthogonal arrays
• Integer programming for classifying orthogonal arrays
• Optimal and efficient semi-Latin squares
• Optimum continuous block designs
• Orthogonal arrays. Theory and applications
• Symmetric factorial designs in blocks
• Theory of optimal designs
• Uniform semi-Latin squares and their Schur-optimality
• Variance and concurrence in block designs, and distance in the corresponding graphs

Cited in

(4)

• Construction of balanced semi-Latin rectangles in block size four: an algorithmic approach
• Constructions for regular-graph semi-Latin rectangles with block size two
• Uniform semi-Latin squares and their Schur-optimality
• On \((M,S)\) and circuit properties of the three optimal \((4\times 4)/4\) semi-Latin squares

Describes a project that uses

Uses Software

• GAP
• GRAPE
• DESIGN
• bliss
• Orthogonal Array

This page was built for publication: Uniform semi-Latin squares and their pairwise-variance aberrations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q830735)