Abstract
The methods of doubling and tripling have been used to construct two-level and three-level symmetrical fractional factorial designs with optimal properties. In this paper, the construction of symmetrical designs is generalized to an asymmetrical case, a novel construction method by amplifying is presented for constructing mixed two- and three-level uniform designs with large run sizes. The analytic relationship between the squared wrap-around \(L_2\)- discrepancy value of the amplified design constructed by amplifying and the wordlength pattern of the initial design is built. Furthermore, the relationships of uniformity and aberration between the amplified design and the corresponding initial design are respectively considered. These results provide a theoretical basis for constructing mixed two- and three-level uniform designs with large run sizes. Finally, some numerical results are provided to support our theoretical results.
Similar content being viewed by others
References
Chatterjee K, Fang KT, Qin H (2005) Uniformity in factorial designs with mixed levels. J Statist Plann Inference 128:593–607
Chen H, Cheng CS (2006) Doubling and projection: a method of constructing two-level designs of resolution IV. Ann Statist 34:546–558
Cheng Y, Zhang RC (2010) On construction of general minimum lower order confounding \(2^{n-m}\) designs with \(n/4+1\le n\le 9n/32\). J Statist Plann Inference 140:2384–2394
Fang KT, Li RZ, Sudjianto A (2006) Design and modeling for computer experiments. Chapman Hall/CRC, New York
Fang KT, Liu MQ, Qin H, Zhou YD (2018) Theory and application of uniform experimental designs. Springer, Singapore
Hickernell FJ (1998) A generalized discrepancy and quadrature error bound. Math Comp 67:299–322
Li HY, Qin H (2018) Some new results on Triple designs. Statist Probab Lett 139:1–9
Ou ZJ, Qin H (2010) Some applications of indicator function in two-level factorial designs. Statist Probab Lett 80:19–25
Ou ZJ, Qin H (2017) Analytic connections between a double design and its original design in terms of different optimality criteria. Commun Statist Theory Methods 46:7630–7641
Ou ZJ, Zhang MH, Qin H (2019) Tripling of fractional factorial designs. J Statist Plann Inference 199:151–159
Sun FS, Wang YP, Xu H (2019) Uniform projection designs. Ann Statist 47:641–661
Xu H, Wu CFJ (2001) Generalized minimum aberration for asymmetrical fractional factorial designs. Ann Statist 29:1066–1077
Xu H, Cheng CS (2008) A complementary design theory for doubling. Ann Statist 36:445–457
Zhang RC, Cheng Y (2010) General minimum lower order confounding designs: an overview and a construction theory. J Statist Plann Inference 140:1719–1730
Acknowledgements
This work was partially supported by the National Natural Science Foundation of China (Nos. 11701213; 11961027; 11871237); Natural Science Foundation of Hunan Province (No. 2020JJ4497) and Scientific Research Plan Item of Hunan Provincial Department of Education (Nos. 19A403, 18A284).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Li, H., Huang, X., Xue, H. et al. A novel method for constructing mixed two- and three-level uniform factorials with large run sizes. Stat Papers 62, 2907–2921 (2021). https://doi.org/10.1007/s00362-020-01219-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-020-01219-8