A semi-infinite programming based algorithm for finding minimax optimal designs for nonlinear models

Minimax optimal experimental designs are notoriously difficult to study largely because the optimality criterion is not differentiable and there is no effective algorithm for generating them. We apply semi-infinite programming (SIP) to solve minimax design problems for nonlinear models in a systematic way using a discretization based strategy and solvers from the General Algebraic Modeling System (GAMS). Using popular models from the biological sciences, we show our approach produces minimax optimal designs that coincide with the few theoretical and numerical optimal designs in the literature. We also show our method can be readily modified to find standardized maximin optimal designs and minimax optimal designs for more complicated problems, such as when the ranges of plausible values for the model parameters are dependent and we want to find a design to minimize the maximal inefficiency of estimates for the model parameters.

[1]  H. Wynn.  Results in the Theory and Construction of D‐Optimum Experimental Designs , 1972 .

[2]  Berç Rustem,et al.  A global optimization algorithm for generalized semi-infinite, continuous minimax with coupled constraints and bi-level problems , 2009, J. Glob. Optim..

[3]  R. Dennis Cook,et al.  Heteroscedastic G-optimal Designs , 1993 .

[4]  Holger Dette,et al.  Maximin optimal designs for the compartmental model , 2003 .

[5]  Bartosz Kuczewski.  Computational aspects of discrimination between models of dynamic systems , 2006 .

[6]  V. Fedorov,et al.  Convex design theory 1 , 1980 .

[7]  Kenneth O. Kortanek,et al.  Semi-Infinite Programming: Theory, Methods, and Applications , 1993, SIAM Rev..

[8]  H. Luthi.  Algorithms for Worst-Case Design and Applications to Risk Management:Algorithms for Worst-Case Design and Applications to Risk Management , 2002 .

[9]  Armen Der Kiureghian,et al.  Adaptive Approximations and Exact Penalization for the Solution of Generalized Semi-infinite Min-Max Problems , 2003, SIAM J. Optim..

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

[11]  Ilya S. Molchanov,et al.  Steepest descent algorithms in a space of measures , 2002, Stat. Comput..

[12]  Marco A. López,et al.  Semi-infinite programming , 2007, Eur. J. Oper. Res..

[13]  William K. W. Li.  Consideration of errors in estimating kinetic parameters based on Michaelis‐Menten formalism in microbial ecology , 1983 .

[14]  Wilfried Seidel,et al.  A minimax algorithm for constructing optimal symmetrical balanced designs for a logistic regression model , 2000 .

[15]  Timothy E. O'Brien,et al.  Designing for Parameter Subsets in Gaussian Nonlinear Regression Models , 2021, Journal of Data Science.

[16]  D. G. Watts,et al.  A Quadratic Design Criterion for Precise Estimation in Nonlinear Regression Models , 1985 .

[17]  J. E. Falk,et al.  Infinitely constrained optimization problems , 1976 .

[18]  Arne Drud,et al.  CONOPT: A GRG code for large sparse dynamic nonlinear optimization problems , 1985, Math. Program..

[19]  K. Kortanek,et al.  Equivalence Theorems and Cutting Plane Algorithms for a Class of Experimental Design Problems , 1977 .

[20]  Holger Dette,et al.  On the number of support points of maximin and Bayesian optimal designs , 2007, 0708.1901.

[21]  Toby J. Mitchell,et al.  An algorithm for the construction of “ D -optimal” experimental designs , 2000 .

[22]  Douglas C. Montgomery,et al.  An Expository Paper on Optimal Design , 2011 .

[23]  W. Wong,et al.  Minimax d-optimal designs for item response theory models , 2000 .

[24]  K. Chaloner,et al.  Optimal Bayesian design applied to logistic regression experiments , 1989 .

[25]  Jesús López-Fidalgo,et al.  MV-optimization in Simple Linear Regression , 1995 .

[26]  Christine H. Müller,et al.  Maximin efficient designs for estimating nonlinear aspects in linear models , 1995 .

[27]  K. Chaloner,et al.  Bayesian Experimental Design: A Review , 1995 .

[28]  R. Prentice,et al.  A generalization of the probit and logit methods for dose response curves. , 1976, Biometrics.

[29]  Weichung Wang,et al.  Minimax optimal designs via particle swarm optimization methods , 2015, Stat. Comput..

[30]  Maciej Patan,et al.  Optimum Design of Experiments for Enzyme Inhibition Kinetic Models , 2011, Journal of biopharmaceutical statistics.

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

[32]  John C. Plummer,et al.  A Multistart Scatter Search Heuristic for Smooth NLP and MINLP Problems , 2005 .

[33]  Holger Dette,et al.  Designing Experiments with Respect to ‘Standardized’ Optimality Criteria , 1997 .

[34]  Andrej Pázman,et al.  Foundations of Optimum Experimental Design , 1986 .

[35]  Oliver Stein,et al.  On generalized semi-infinite optimization and bilevel optimization , 2002, Eur. J. Oper. Res..

[36]  Holger Dette,et al.  Bayesian D-optimal designs for exponential regression models , 1997 .

[37]  Berç Rustem,et al.  Semi-Infinite Programming and Applications to Minimax Problems , 2003, Ann. Oper. Res..

[38]  Guillaume Sagnol,et al.  Computing Optimal Designs of multiresponse Experiments reduces to Second-Order Cone Programming , 2009, 0912.5467.

[39]  Shan Sun,et al.  A class of adaptive distribution-free procedures , 1997 .

[40]  Holger Dette,et al.  E-optimal designs for the Michaelis–Menten model , 1999 .

[41]  Holger Dette,et al.  MINIMAX OPTIMAL DESIGNS IN NONLINEAR REGRESSION MODELS , 1998 .

[42]  Weng Kee Wong,et al.  An Introduction to Optimal Designs for Social and Biomedical Research , 2009 .

[43]  Weng Kee Wong,et al.  Optimal designs for the power logistic model , 2004 .

[44]  Ray-Bing Chen,et al.  Optimal minimax designs over a prespecified interval in a heteroscedastic polynomial model. , 2008, Statistics & probability letters.

[45]  D'avid Papp,et al.  Optimal Designs for Rational Function Regression , 2010, 1009.1444.

[46]  W K Wong,et al.  Minimax D‐Optimal Designs for the Logistic Model , 2000, Biometrics.

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

[48]  J. Kiefer,et al.  Time- and Space-Saving Computer Methods, Related to Mitchell's DETMAX, for Finding D-Optimum Designs , 1980 .

[49]  W. J. Hill,et al.  Design of Experiments for Subsets of Parameters , 1974 .

[50]  Weng Kee Wong,et al.  A unified approach to the construction of minimax designs , 1992 .

[51]  Ying Zhang,et al.  Bayesian D-Optimal Design for Generalized Linear Models , 2006 .

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

[53]  R. Sitter.  Robust designs for binary data , 1992 .