Folded over non-orthogonal designs. (English) Zbl 1140.62064
Summary: We use the notion of minimal dependent sets (MDS) to introduce MDS-resolution and MDS-aberration as criteria for comparing non-orthogonal foldover designs, and discuss the ideas and their usefulness. We also develop a fast isomorphism check that uses a cyclic matrix defined on the design before it is folded over. By doing so, the speed of the check for comparing two isomorphic designs is increased relative to merely applying an isomorphism check to the foldover design. This relative difference becomes greater as the design size increases. Finally, we use the isomorphism check to obtain a catalog of minimum MDS-aberration designs for some useful \(n\) and \(k\) and discuss an algorithm for obtaining “good” larger designs.
MSC:
| 62K15 | Factorial statistical designs |
| 62Q05 | Statistical tables |
| 62J05 | Linear regression; mixed models |
Keywords:
aberration; fractional factorial; Hadamard matrix; minimal dependent sets; Plackett-Burman designs; resolution; isomorphism; equivalenceReferences:
| [1] | Banerjee, K. S.; Federer, W. T., On a special subset giving irregular fractional replicate of a \(2^n\) factorial experiment, J. Roy. Statist. Soc. Ser. B, 29, 292-299 (1967) |
| [2] | Bingham, D. R.; Chipman, H. A., Incorporating prior information in optimal design for model selection, Technometrics, 49, 155-163 (2007) |
| [3] | Clark, J. B.; Dean, A. M., Equivalence of fractional factorial designs, Statist. Sinica, 11, 537-547 (2001) · Zbl 0980.62058 |
| [4] | John, P. W.M., Three-quarter replicates of \(2^n\) designs, Biometrics, 18, 172-184 (1962) · Zbl 0105.12101 |
| [5] | John, P. W.M., Blocking of \(3(2^{n - k})\) designs, Technometrics, 6, 371-376 (1964) · Zbl 0124.10901 |
| [6] | Jones, B. A.; Li, W.; Nachtsheim, C. J.; Ye, K. Q., Model discrimination—another perspective on model-robust designs, J. Statist. Plann. Inference, 137, 1576-1583 (2007) · Zbl 1110.62103 |
| [7] | Lin, C.D., Sitter, R.R., 2007. An isomorphism check for two-level fractional factorial designs. J. Statist. Plann. Inference, in press.; Lin, C.D., Sitter, R.R., 2007. An isomorphism check for two-level fractional factorial designs. J. Statist. Plann. Inference, in press. · Zbl 1130.62078 |
| [8] | Loeppky, J.; Sitter, R. R.; Tang, B., Non-regular designs with desirable projection properties, Technometrics, 49, 454-467 (2007) |
| [9] | Margolin, B. H., Results on factorial designs of resolution IV for the \(2^n\) and \(2^n 3^m\) series, Technometrics, 11, 431-444 (1969) · Zbl 0179.24002 |
| [10] | Miller, A.; Sitter, R. R., Using the folded-over 12-run Plackett-Burman design to consider interactions, Technometrics, 43, 44-55 (2001) · Zbl 1072.62622 |
| [11] | Miller, A.; Sitter, R. R., Choosing columns from the 12-run Plackett-Burman design, Statist. Probab. Lett., 67, 193-201 (2004) · Zbl 1059.62084 |
| [12] | Miller, A.; Sitter, R. R., Using folded over non-orthogonal designs, Technometrics, 47, 502-513 (2005) |
| [13] | Plackett, R. L.; Burman, J. P., The design of optimum multifactorial experiments, Biometrika, 33, 305-325 (1946) · Zbl 0063.06274 |
| [14] | Schott, J. R., Matrix Analysis for Statistics (1997), Wiley: Wiley New York, NY · Zbl 0872.15002 |
| [15] | Srivastava, J. N., Designs for searching non-negligible effects, (A Survey of Statistical Design and Linear Models (1975), North-Holland: North-Holland Amsterdam) · Zbl 0313.62060 |
| [16] | Webb, S., Non-orthogonal designs of even resolution, Technometrics, 10, 291-299 (1968) · Zbl 0174.22501 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
