ABSTRACT

This article considers the problem of selecting two-level designs under the baseline parameterisation when some two-factor interactions are important. We propose a minimum aberration criterion, which minimises the bias caused by the non-negligible effects. Using this criterion, a class of optimal designs can be further distinguished from one another, and we present an algorithm to find the minimum aberration designs among the D-optimal designs. Sixteen-run and twenty-run designs are summarised for practical use.

References

1. Banerjee, T., & Mukerjee, R. (2008). Optimal factorial designs for cDNA microarray experiments. The Annals of Applied Statistics, 2(1), 366–385. https://doi.org/10.1214/07-AOAS144 [Crossref], [Web of Science ®], [Google Scholar]
2. Cheng, C. S. (1980). Orthogonal arrays with variable numbers of symbols. The Annals of Statistics, 8(2), 447–453. https://doi.org/10.1214/aos/1176344964 [Crossref], [Web of Science ®], [Google Scholar]
3. Fries, A., & Hunter, W. G. (1980). Minimum aberration 2^k−p designs. Technometrics, 22(4), 601–608. https://doi.org/10.1080/00401706.1980.10486210 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
4. Ke, W., & Tang, B. (2003). Selecting 2m−p2m−p designs using a minimum aberration criterion when some two-factor interactions are important. Technometrics, 45(4), 352–360. https://doi.org/10.1198/004017003000000186 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
5. Kerr, K. F. (2006). Efficient 2k2k factorial designs for blocks of size 2 with microarray applications. Journal of Quality Technology, 38(4), 309–318. https://doi.org/10.1080/00224065.2006.11918620 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
6. Li, P., Miller, A., & Tang, B. (2014). Algorithmic search for baseline minimum aberration designs. Journal of Statistical Planning and Inference, 149, 172–182. https://doi.org/10.1016/j.jspi.2014.02.009 [Crossref], [Web of Science ®], [Google Scholar]
7. Mukerjee, R., & Tang, B. (2012). Optimal fractions of two-level factorials under a baseline parameterization. Biometrika, 99(1), 71–84. https://doi.org/10.1093/biomet/asr071 [Crossref], [Web of Science ®], [Google Scholar]
8. Sun, D. X., Li, W., & Ye, K. Q. (2008). Algorithmic construction of catalogs of non-isomorphic two-level orthogonal designs for economic run sizes. Statistics and Applications, 6, 141–155. [Google Scholar]
9. Sun, C. Y., & Tang, B. (2020). Relationship between orthogonal and baseline parameterizations and its applications to design constructions. Statistica Sinica. Accepted. [Google Scholar]
10. Tang, B., & Deng, L. Y. (1999). Minimum G2G2-aberration for nonregular fractional factorial designs. Annals of Statistics, 27(6), 1914–1926. https://doi.org/10.1214/aos/1017939244 [Web of Science ®], [Google Scholar]
11. Wu, C. J., & Hamada, M. S. (2011). Experiments: planning, analysis, and optimization (Vol. 552). John Wiley & Sons. [Google Scholar]
12. Yang, Y. H., & Speed, T. (2002). Design issues for cDNA microarray experiments. Nature Reviews Genetics, 3(8), 579–588. https://doi.org/10.1038/nrg863 [Crossref], [Web of Science ®], [Google Scholar]