Exchange Algorithms for Constructing Model-Robust Experimental Designs

Optimal experimental design procedures, utilizing criteria such as D-optimality, are useful for producing designs for quantitative responses, often under nonstandard conditions such as constrained design spaces. However, these methods require a priori knowledge of the exact form of the response function, an often unrealistic assumption. Model-robust designs are those that, from our perspective, are efficient with respect to a set of possible models. In this paper, we develop a model-robust technique motivated by a connection to multiresponse D-optimal design. This link spawns a generalization of the modified Fedorov exchange algorithm, which is then used to construct exact model-robust designs. We also study the effectiveness of designs robust for a small set of models compared with designs that account for much larger sets. We give several examples and compare our designs with two model-robust procedures in the literature.

[1]  G. Piepel.  Programs for Generating Extreme Vertices and Centroids of Linearly Constrained Experimental Regions , 1988 .

[2]  W. DuMouchel,et al.  A simple Bayesian modification of D-optimal designs to reduce dependence on an assumed model , 1994 .

[3]  D. Wiens,et al.  Integer-Valued, Minimax Robust Designs for Estimation and Extrapolation in Heteroscedastic, Approximately Linear Models , 2000 .

[4]  Grace Montepiedra,et al.  Minimum bias designs with constraints , 1997 .

[5]  J. Cornell.  Experiments with Mixtures: Designs, Models and the Analysis of Mixture Data , 1982 .

[6]  M. E. Johnson,et al.  Some Guidelines for Constructing Exact D-Optimal Designs on Convex Design Spaces , 1983 .

[7]  Connie M. Borror,et al.  Model-Robust Optimal Designs: A Genetic Algorithm Approach , 2004 .

[8]  Lynne B. Hare,et al.  Experiments with Mixtures: Designs, Models and the Analysis of Mixture Data, 2nd Ed. , 1991 .

[9]  Christopher J. Nachtsheim,et al.  Model Robust, Linear-Optimal Designs , 1982 .

[10]  Hugh A. Chipman,et al.  Incorporating Prior Information in Optimal Design for Model Selection , 2007, Technometrics.

[11]  R. D. Cook,et al.  A Comparison of Algorithms for Constructing Exact D-Optimal Designs , 1980 .

[12]  Gregory F. Piepel,et al.  Construction of a 21-Component Layered Mixture Experiment Design Using a New Mixture Coordinate-Exchange Algorithm , 2005 .

[13]  James R. Schott,et al.  Matrix Analysis for Statistics , 2005 .

[14]  R. Schwabe,et al.  THE REDUCTION OF DESIGN PROBLEMS FOR MULTIVARIATE EXPERIMENTS TO UNIVARIATE POSSIBILITIES AND THEIR LIMITATIONS , 1996 .

[15]  W. Näther.  Optimum experimental designs , 1994 .

[16]  Julie Zhou,et al.  D-optimal minimax regression designs on discrete design space , 2008 .

[17]  G. Box,et al.  A Basis for the Selection of a Response Surface Design , 1959 .

[18]  Holger Dette,et al.  Robust designs for polynomial regression by maximizing a minimum of D- and D1-efficiencies , 2001 .

[19]  H. Dette,et al.  A Generalization of $D$- and $D_1$-Optimal Designs in Polynomial Regression , 1990 .

[20]  W. Welch.  A mean squared error criterion for the design of experiments , 1983 .

[21]  M. Deaton,et al.  Response Surfaces: Designs and Analyses , 1989 .

[22]  Dennis K. J. Lin,et al.  Bayesian D-optimal supersaturated designs , 2008 .

[23]  Abbas Seifi,et al.  Optimal design of multi-response experiments using semi-definite programming , 2009 .

[24]  W. J. Studden,et al.  Theory Of Optimal Experiments , 1972 .

[25]  S. CHANG,et al.  An algorithm to generate near D-optimal designs for multiple response surface models , 1997 .

[26]  R. K. Meyer,et al.  The Coordinate-Exchange Algorithm for Constructing Exact Optimal Experimental Designs , 1995 .

[27]  Enrique del Castillo,et al.  Model-Robust Process Optimization Using Bayesian Model Averaging , 2005, Technometrics.

[28]  Anthony C. Atkinson,et al.  Optimum Experimental Designs, with SAS , 2007 .

[29]  Kenny Q. Ye,et al.  Model-robust supersaturated and partially supersaturated designs , 2009 .

[30]  Peter Goos,et al.  Model-robust and model-sensitive designs , 2005, Comput. Stat. Data Anal..

[31]  William Li,et al.  Model-Robust Factorial Designs , 2000, Technometrics.

[32]  N. Schaumberger.  Generalization , 1989, Whitehead and Philosophy of Education.

[33]  Toby J. Mitchell,et al.  An Algorithm for the Construction of “D-Optimal” Experimental Designs , 2000, Technometrics.

[34]  Angela R. Neff,et al.  Bayesian Two Stage Design Under Model Uncertainty , 1997 .

[35]  Douglas P. Wiens,et al.  Robust regression designs for approximate polynomial models , 2003 .

[36]  W. Bischoff.  On D-optimal designs for linear models under correlated observations with an application to a linear model with multiple response , 1993 .

[37]  J. Mcginity,et al.  Influence of nonionic surfactants on the physical and chemical properties of a biodegradable psuedolatex , 1994 .

[38]  Harrison M. Wadsworth,et al.  Journal of Quality Technology, The , 2005 .

[39]  Douglas C. Montgomery,et al.  OPTIMIZATION OF A HOUSEHOLD PRODUCT FORMULATION USING A MIXTURE EXPERIMENT , 1995 .

[40]  E. Läuter.  Experimental design in a class of models , 1974 .

[41]  Peter C. M. Molenaar,et al.  Optimal measurement conditions for spatiotemporal eeg/meg source analysis , 2002 .