A two-stage variable selection strategy for supersaturated designs with multiple responses

Abstract

A supersaturated design (SSD), whose run size is not enough for estimating all the main effects, is commonly used in screening experiments. It offers a potential useful tool to investigate a large number of factors with only a few experimental runs. The associated analysis methods have been proposed by many authors to identify active effects in situations where only one response is considered. However, there are often situations where two or more responses are observed simultaneously in one screening experiment, and the analysis of SSDs with multiple responses is thus needed. In this paper, we propose a two-stage variable selection strategy, called the multivariate partial least squares-stepwise regression (MPLS-SR) method, which uses the multivariate partial least squares regression in conjunction with the stepwise regression procedure to select true active effects in SSDs with multiple responses. Simulation studies show that the MPLS-SR method performs pretty good and is easy to understand and implement.

Keywords

Multivariate partial least squares (MPLS) / supersaturated design (SSD) / stepwise regression / variable selection / variable importance in projection

References

[1] Bastien P, Vinzi V E, Tenenhaus M. PLS generalized linear regression. Comput Statist Data Anal, 2005, 48(1): 17-46

[2] Beattie S D, Fong D K H, Lin D K J. A two-stage Bayesian model selection strategy for supersaturated designs. Technometrics, 2002, 44(1): 55-63

[3] Booth K H V, Cox D R. Some systematic supersaturated designs. Technometrics, 1962, 4(4): 489-495

[4] Box G, Meyer R. An analysis for unreplicated fractional factorials. Technometrics, 1986, 28(1): 11-18

[5] Butler N A, Denham M C. The peculiar shrinkage properties of partial least squares regression. J Roy Statist Soc Ser B, 2000, 62(3): 585-593

[6] Chipman H, Hamada H, Wu C F J. A Bayesian variable-selection approach for analyzing designed experiments with complex aliasing. Technometrics, 1997, 39(4): 372-381

[7] Fang K T, Lin D K J, Liu M Q. Optimal mixed-level supersaturated design. Metrika, 2003, 58(3): 279-291

[8] Garthwaite P H. An interpretation of partial least squares. J Amer Statist Assoc, 1994, 89(425): 122-127

[9] Georgiou S D. Modelling by supersaturated designs. Comput Statist Data Anal, 2008, 53(2): 428-435

[10] Helland I S. Partial least squares regression and statistical models. Scand J Statist, 1990, 17(2): 97-114

[11] Holcomb D R, Montgomery D C, Carlyle W M. An alysis of supersaturated designs. J Quality Technol, 2003, 35(1): 13-27

[12] Höoskuldsson A. Variable and subset selection in PLS regression. Chemometrics & Intelligent Laboratory Systems, 2001, 55: 23-38

[13] Li P, Zhao S L, Zhang R C. A cluster analysis selection strategy for supersaturated designs. Comput Statist Data Anal, 2010, 54: 1605-1612

[14] Li R, Lin D K J. Data analysis of supersaturated designs. Statist Probab Lett, 2002, 59(2): 135-144

[15] Li R, Lin D K J. Analysis methods for supersaturated design: some comparisons. J Data Sci, 2003, 1(3): 249-260

[16] Lin D K J. A new class of supersaturated designs. Technometrics, 1993, 35(1): 28-31

[17] Liu Y, Liu M Q. Construction of optimal supersaturated design with large number of levels. J Statist Plann Inference, 2011, 141: 2035-2043

[18] Liu Y, Liu M Q. Construction of equidistant and weak equidistant supersaturated designs. Metrika, 2012, 75(1): 33-53

[19] Lu X, Wu X. A strategy of searching active factors in supersaturated screening experiments. J Quality Technol, 2004, 36(4): 392-399

[20] Marley C J, Woods D C. A comparison of design and model selection methods for supersaturated experiments. Comput Statist Data Anal, 2010, 54(12): 3158-3167

[21] Phoa F K H, Pan Y H, Xu H. Analysis of supersaturated designs via the Dantzig selector. J Statist Plann Inference, 2009, 139: 2362-2372

[22] Sun F S, Lin D K J, Liu M Q. On construction of optimal mixed-level supersaturated designs. Ann Statist, 2011, 39(2): 1310-1333

[23] Wang H W. Partial Least-squares Regression Method and Applications. Beijing: National Defence Industry Press, 1999 (in Chinese)

[24] Westfall P H, Young S S, Lin D K J. Forward selection error control in the analysis of supersaturated design. Statist Sinica, 1998, 8(1): 101-117

[25] Williams K R. Designed experiments. Rubber Age, 1968, 100: 65-71

[26] Zhang Q Z, Zhang R C, Liu M Q. A method for screening active effects in supersaturated designs. J Statist Plann Inference, 2007, 137(6): 2068-2079

RIGHTS & PERMISSIONS

2014 Higher Education Press and Springer-Verlag Berlin Heidelberg