A Metaheuristic Adaptive Cubature Based Algorithm to Find Bayesian Optimal Designs for Nonlinear Models

Abstract Finding Bayesian optimal designs for nonlinear models is a difficult task because the optimality criterion typically requires us to evaluate complex integrals before we perform a constrained optimization. We propose a hybridized method where we combine an adaptive multidimensional integration algorithm and a metaheuristic algorithm called imperialist competitive algorithm to find Bayesian optimal designs. We apply our numerical method to a few challenging design problems to demonstrate its efficiency. They include finding D-optimal designs for an item response model commonly used in education, Bayesian optimal designs for survival models, and Bayesian optimal designs for a four-parameter sigmoid Emax dose response model. Supplementary materials for this article are available online and they contain an R package for implementing the proposed algorithm and codes for reproducing all the results in this article.

[1]  Xin-She Yang,et al.  Metaheuristic Optimization: Algorithm Analysis and Open Problems , 2011, SEA.

[2]  A. Genz,et al.  Computation of Multivariate Normal and t Probabilities , 2009 .

[3]  Tim B. Swartz,et al.  Approximating Integrals Via Monte Carlo and Deterministic Methods , 2000 .

[4]  Xin-She Yang,et al.  Engineering Optimization: An Introduction with Metaheuristic Applications , 2010 .

[5]  Kay Chen Tan,et al.  Finding High-Dimensional D-Optimal Designs for Logistic Models via Differential Evolution , 2019, IEEE Access.

[6]  David Kendrick,et al.  GAMS, a user's guide , 1988, SGNM.

[7]  J. Kiefer,et al.  Optimum Designs in Regression Problems , 1959 .

[8]  John Hinde,et al.  On Bayesian D‐optimum Design Criteria and the Equivalence Theorem in Non‐linear Models , 1997 .

[9]  David M. Steinberg,et al.  Fast Computation of Designs Robust to Parameter Uncertainty for Nonlinear Settings , 2009, Technometrics.

[10]  Weichung Wang,et al.  Using animal instincts to design efficient biomedical studies via particle swarm optimization , 2014, Swarm Evol. Comput..

[11]  C. Dunnett,et al.  A BIVARIATE GENERALIZATION OF STUDENT'S t-DISTRIBUTION, WITH TABLES FOR CERTAIN SPECIAL CASES , 1954 .

[12]  Arnold Neumaier,et al.  Introduction to Numerical Analysis , 2001 .

[13]  A. Atkinson.  The Usefulness of Optimum Experimental Designs , 1996 .

[14]  Paul Van Dooren,et al.  An adaptive algorithm for numerical integration over the n-cube , 1976 .

[15]  W. G. Hunter,et al.  The Experimental Study of Physical Mechanisms , 1965 .

[16]  I. Olkin,et al.  Collected Papers III , 1985 .

[17]  Jack Kiefer,et al.  Collected Papers III: Design of Experiments , 1984 .

[18]  Adaptive optimal designs for dose-finding studies based on sigmoid Emax models , 2014 .

[19]  Hung-Yi Lu,et al.  Application of Optimal Designs to Item Calibration , 2014, PloS one.

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

[21]  G. H. Fischer,et al.  Remarks on “equivalent linear logistic test models” by Bechger, Verstralen, and Verhelst (2002) , 2004 .

[22]  F. Hsuan,et al.  Adaptive Designs for Dose-Finding Studies Based on Sigmoid E max Model , 2007, Journal of biopharmaceutical statistics.

[23]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[24]  R. Schwabe,et al.  On optimal designs for censored data , 2015 .

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

[26]  H. Chernoff.  Locally Optimal Designs for Estimating Parameters , 1953 .

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

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

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

[30]  Weng Kee Wong,et al.  A semi-infinite programming based algorithm for finding minimax optimal designs for nonlinear models , 2014, Stat. Comput..

[31]  Valéria Lima Passos,et al.  Maximin Calibration Designs for the Nominal Response Model: an Empirical Evaluation , 2004 .

[32]  A. Azzalini,et al.  Statistical applications of the multivariate skew normal distribution , 2009, 0911.2093.

[33]  Terje O. Espelid,et al.  Algorithm 698: DCUHRE: an adaptive multidemensional integration routine for a vector of integrals , 1991, TOMS.

[34]  Terje O. Espelid,et al.  An adaptive algorithm for the approximate calculation of multiple integrals , 1991, TOMS.

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

[36]  F. Pukelsheim,et al.  Efficient rounding of approximate designs , 1992 .

[37]  Abhyuday Mandal,et al.  d-QPSO: A Quantum-Behaved Particle Swarm Technique for Finding D-Optimal Designs With Discrete and Continuous Factors and a Binary Response , 2018, Technometrics.

[38]  Heinz Holling,et al.  Application of imperialist competitive algorithm to find minimax and standardized maximin optimal designs , 2017, Comput. Stat. Data Anal..

[39]  James McGree,et al.  Probability-based optimal design , 2008 .

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

[41]  J. Macdougall.  Analysis of Dose–Response Studies—E max Model , 2006 .

[42]  Heinz Holling,et al.  An Introduction to Optimal Design Some Basic Issues Using Examples From Dyscalculia Research , 2013 .

[43]  Alan Genz,et al.  An adaptive algorithm for numerical integration over an n-dimensional rectangular region , 1980 .

[44]  Weng Kee Wong,et al.  Model-based optimal design of experiments - semidefinite and nonlinear programming formulations. , 2016, Chemometrics and intelligent laboratory systems : an international journal sponsored by the Chemometrics Society.

[45]  Anthony C. Atkinson,et al.  DT-optimum designs for model discrimination and parameter estimation , 2008 .

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

[47]  J. Rost.  The Growing Family of Rasch Models , 2001 .

[48]  Caro Lucas,et al.  Imperialist competitive algorithm: An algorithm for optimization inspired by imperialistic competition , 2007, 2007 IEEE Congress on Evolutionary Computation.

[49]  P. Goos,et al.  Quadrature Methods for Bayesian Optimal Design of Experiments With Nonnormal Prior Distributions , 2018 .

[50]  Corwin L. Atwood,et al.  Optimal and Efficient Designs of Experiments , 1969 .

[51]  W. Wong,et al.  Standardized maximim D-optimal designs for enzyme kinetic inhibition models. , 2017, Chemometrics and intelligent laboratory systems : an international journal sponsored by the Chemometrics Society.

[52]  H. Holling,et al.  Task difficulty prediction of figural analogies , 2016 .

[53]  Maria Konstantinou,et al.  Optimal designs for two-parameter nonlinear models with application to survival models , 2014 .