D-optimal designs via a cocktail algorithm

A fast new algorithm is proposed for numerical computation of (approximate) D-optimal designs. This cocktail algorithm extends the well-known vertex direction method (VDM; Fedorov in Theory of Optimal Experiments, 1972) and the multiplicative algorithm (Silvey et al. in Commun. Stat. Theory Methods 14:1379–1389, 1978), and shares their simplicity and monotonic convergence properties. Numerical examples show that the cocktail algorithm can lead to dramatically improved speed, sometimes by orders of magnitude, relative to either the multiplicative algorithm or the vertex exchange method (a variant of VDM). Key to the improved speed is a new nearest neighbor exchange strategy, which acts locally and complements the global effect of the multiplicative algorithm. Possible extensions to related problems such as nonparametric maximum likelihood estimation are mentioned.

[1]  Yaming Yu,et al.  Strict monotonicity and convergence rate of Titterington's algorithm for computing D-optimal designs , 2010, Comput. Stat. Data Anal..

[2]  S. Mandal,et al.  Two classes of multiplicative algorithms for constructing optimizing distributions , 2006, Comput. Stat. Data Anal..

[3]  Holger Dette,et al.  Locally D-optimal Designs for Exponential Regression , 2004 .

[4]  Jennifer Seberry,et al.  D-optimal designs , 2011 .

[5]  K. Chaloner,et al.  Optimum experimental designs for properties of a compartmental model. , 1993, Biometrics.

[6]  Yong Wang,et al.  Dimension-reduced nonparametric maximum likelihood computation for interval-censored data , 2008, Comput. Stat. Data Anal..

[7]  Geurt Jongbloed,et al.  The Iterative Convex Minorant Algorithm for Nonparametric Estimation , 1998 .

[8]  Changbao Wu,et al.  Some Algorithmic Aspects of the Theory of Optimal Designs , 1978 .

[9]  Jon A. Wellner,et al.  A Hybrid Algorithm for Computation of the Nonparametric Maximum Likelihood Estimator from Censored Data , 1997 .

[10]  Michael Jackson,et al.  Optimal Design of Experiments , 1994 .

[11]  Linda M. Haines.  A class of equivalent problems in statistics and operational research , 1998 .

[12]  J. Kiefer.  General Equivalence Theory for Optimum Designs (Approximate Theory) , 1974 .

[13]  Holger Dette,et al.  Improving updating rules in multiplicative algorithms for computing D-optimal designs , 2008, Comput. Stat. Data Anal..

[14]  B. Lindsay.  The Geometry of Mixture Likelihoods: A General Theory , 1983 .

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

[16]  Yaming Yu.  Monotonic convergence of a general algorithm for computing optimal designs , 2009, 0905.2646.

[17]  Saumen Mandal,et al.  Construction of optimal designs using a clustering approach , 2006 .

[18]  Anthony C. Atkinson,et al.  Optimum Experimental Designs , 1992 .

[19]  J. Kiefer,et al.  The Equivalence of Two Extremum Problems , 1960, Canadian Journal of Mathematics.

[20]  Luc Pronzato,et al.  Improvements on removing nonoptimal support points in D-optimum design algorithms , 2007, 0706.4394.

[21]  Yaming Yu,et al.  Improved EM for Mixture Proportions with Applications to Nonparametric ML Estimation for Censored Data , 2010, 1002.3640.

[22]  S. Silvey,et al.  An algorithm for optimal designs on a design space , 1978 .

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

[24]  D. Titterington.  Estimation of Correlation Coefficients by Ellipsoidal Trimming , 1978 .

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

[26]  Dankmar Böhning,et al.  A vertex-exchange-method in D-optimal design theory , 1986 .

[27]  Holger Dette,et al.  Locally E-optimal designs for expo-nential regression models , 2003 .

[28]  Corwin L. Atwood,et al.  Convergent Design Sequences, for Sufficiently Regular Optimality Criteria , 1976 .

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