Abstract
Uniform designs have been widely used in physical and computer experiments due to their robust performances. The level permutation method can efficiently construct uniform designs with both lower discrepancy and less aberration. However, the related existing literature has mostly discussed uniform fixed-level designs, the construction of uniform mixed-level designs has been quite few studied. In this paper, a novel level permutation method for constructing uniform mixed-level designs is proposed. Our main idea is to perform level permutations on a new class of designs, called minimum average discrepancy designs, rather than generalized minimum aberration designs as in the fixed-level case. Several theoretical results on the design optimality and construction are obtained. Numerical results suggest the good performance of the resulting designs under various popular discrepancies.
Similar content being viewed by others
References
Elsawah AM, Qin H (2016) Asymmetric uniform designs based on mixture discrepancy. J Appl Stat 43:2280–2294
Fang KT (1980) The uniform design: application of number-theoretic methods in experimental design. Acta Math Appl Sin 3:363–372
Fang KT, Lu X, Winker P (2003) Lower bounds for centered and wrap-around \({L}_2\)-discrepancies and construction of uniform designs by threshold accepting. J Complex 19:692–711
Fang KT, Lu X, Tang Y, Yin J (2004) Constructions of uniform designs by using resolvable packings and coverings. Discrete Math 274:25–40
Fang KT, Li R, Sudjianto A (2006a) Design and modeling for computer experiments. Chapman & Hall/CRC, Boca Raton
Fang KT, Maringer D, Tang Y, Winker P (2006b) Lower bounds and stochastic optimization algorithms for uniform designs with three or four levels. Math Comput 75:859–878
Hedayat AS, Sloane NJA, Stufken J (1999) Orthogonal arrays: theory and applications. Springer, New York
Hickernell FJ (1998a) A generalized discrepancy and quadrature error bound. Math Comput 67:299–322
Hickernell FJ (1998b) Lattice rules: how well do they measure up? In: Hellekalek P, Larcher G (eds) Random and quasi-random point sets. Springer, New York, pp 109–166
Hickernell FJ, Liu M (2002) Uniform designs limit aliasing. Biometrika 89:893–904
Liu Y, Liu M (2012) Construction of equidistant and weak equidistant supersaturated designs. Metrika 75:33–53
Mukerjee R, Wu CFJ (1995) On the existence of saturated and nearly saturated asymmetrical orthogonal arrays. Ann Stat 23:2102–2115
Shrikhande S (1964) Generalized Hadamard matrices and orthogonal arrays of strength two. Can J Math 16:736–740
Tang Y, Xu H (2013) An effective construction method for multi-level uniform designs. J Stat Plan Inference 143:1583–1589
Tang B, Zhou J (2009) Existence and construction of two-level orthogonal arrays for estimating main effects and some specified two-factor interactions. Stat Sin 19:1193–1201
Tang Y, Xu H, Lin DKJ (2012) Uniform fractional factorial designs. Ann Stat 40:891–907
Wang JC, Wu CFJ (1991) An approach to the construction of asymmetrical orthogonal arrays. J Am Stat Assoc 86:450–456
Xu H (2003) Minimum moment aberration for nonregular designs and supersaturated designs. Stat Sin 13:691–708
Xu G, Zhang J, Tang Y (2014) Level permutation method for constructing uniform designs under the wrap-around \({L}_2\)-discrepancy. J Complex 30:46–53
Yang F, Zhou Y, Zhang A (2019) Mixed-level column augmented uniform designs. J Complex 53:23–39
Zhang Q, Wang Z, Hu J, Qin H (2015) A new lower bound for wrap-around \(l_2\)-discrepancy on two and three mixed level factorials. Stat Probab Lett 96:133–140
Zhou Y, Xu H (2014) Space-filling fractional factorial designs. J Am Stat Assoc 109:1134–1144
Zhou Y, Fang KT, Ning J (2013) Mixture discrepancy for quasi-random point sets. J Complex 29:283–301
Acknowledgements
The authors sincerely thank the editor, the associate editor, and two referees for their comments, which have led to the improvement of the paper. Bochuan Jiang’s work is partially supported by the Fundamental Research Funds for the Central Universities 2021RC219 and NSFC Grant 12001036. Fei Wang’s work is partially supported by NSFC Grants 12071014 and 12131001, SSFC Grant 19ZDA121 and LMEQF. Yaping Wang’s work is partially supported by NSFC Grants 11901199 and 71931004 and Shanghai “Chenguang Program” 19CG26.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interests
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Proof of Theorem 1
For any row \({\varvec{d}}_i=(d_{i,1},\ldots ,d_{i,n})\) of D, when all level permutations are considered, each n-tuple in \(R_{s_1}^{n_1}\times \cdots \times R_{s_g}^{n_g}\) occurs \(\prod _{k=1}^g[(s_k-1)!]^{n_k}\) times. Then
which is a constant. Similarly,
is also a constant.
For any two distinct rows \({\varvec{d}}_{i_1}\) and \({\varvec{d}}_{i_2}\) of D, when all level permutations of D are considered, each identical pair (z, z) of the \(n_k\) factors with \(s_k\) levels occurs \((s_k-1)!\) times in the \(\delta _{i_1,i_2,k}(D)\) positions of coincidence and each distinct pair \((z_1,z_2)\) occurs \((s_k-2)!\) times in the other \(n_k-\delta _{i_1,i_2,k}(D)\) positions. Then
The desired result follows by combining (3), (5), (10), (11) and (12). \(\square \)
Proof of Theorem 4
Let \(V=R_{s_1}^{n_1}\times \cdots \times R_{s_g}^{n_g}\). For the \((N,s_1^{n_1}\cdots s_g^{n_g})\)-design D without repeated points, let \(m(l_1,\ldots ,l_n)\) denote the number of times that the point \((l_1,\ldots ,l_n)\) occurs in D. Then the design D can be uniquely determined by the following column vector of length \(N_0=\prod _{k=1}^gs_k^{n_k}\), i.e.,
where all points \((l_1,\ldots ,l_n)\) in V are arranged in the lexicographical order. The vector \(\mathbf{y}_D\) is called the frequency vector of the design D. In particular, a full factorial \((N_0,s_1^{n_1}\cdots s_g^{n_g})\)-design has \(\mathbf{y}_D= {\varvec{1}}_{N_0}\), the \(N_0\)-vector of ones. By noticing
the quadratic form of \({\bar{\phi }}(D)\) in terms of \(\mathbf{y}_D\) is obtained as
where for \(i,j=1,\ldots ,s_k\) and \(k=1,\ldots ,g\),
\(\mathbf{A}_k=(a_{i,j}^{(k)})_{s_k\times s_k}\), \(a_{i,j}^{(k)}\) equals \(\gamma (s_k)\) if \(i=j\) and 1 otherwise, \(c_0\), \(\gamma (\cdot )\), \(\alpha (\cdot )\) and \(\beta (\cdot )\) are defined in Theorem 1, \(\mathbf{A}\otimes \mathbf{B}\) denotes the Kronecker product of two matrices \(\mathbf{A}\) and \(\mathbf{B}\). Similarly,
where \(\mathbf{y}_{D^c}\) is the frequency vector of the design \(D^c\). Since \(D^c\) is the complementary design of D, we have
By noting that \(\mathbf{A}_k\mathbf{1}_{s_k}=\gamma (s_k)+s_k-1\), we also obtain
Combining (14), (15), (16) and (17), the desired result follows. \(\square \)
Rights and permissions
About this article
Cite this article
Jiang, B., Wang, F. & Wang, Y. Construction of uniform mixed-level designs through level permutations. Metrika 85, 753–770 (2022). https://doi.org/10.1007/s00184-021-00850-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00184-021-00850-1