论文信息 - Removing non-optimal support points in D-optimum design algorithms

Removing non-optimal support points in D-optimum design algorithms

We show that points can be removed from the design space during the search for a D-optimum design, using a simple inequality satisfied by support points of a D-optimum design measure. This inequality does not seem to have been noticed before but may be used for accelerating algorithms for D-optimum design, or for the determination of minimum covering ellipsoids, as used, e.g., in ellipsoidal trimming.

Luc Pronzato | L. Pronzato

[1] D. Böhning. Likelihood inference for mixtures: Geometrical and other constructions of monotone step-length algorithms , 1989 .

[2] Douglas M. Hawkins,et al. Exact iterative computation of the robust multivariate minimum volume ellipsoid estimator , 1993 .

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

[4] Steven E. Rigdon,et al. Model-Oriented Design of Experiments , 1997, Technometrics.

[5] D. Titterington. Optimal design: Some geometrical aspects of D-optimality , 1975 .

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

[7] P. L. Davies,et al. The asymptotics of Rousseeuw's minimum volume ellipsoid estimator , 1992 .

[8] B. Torsney. A Moment Inequality and Monotonicity of an Algorithm , 1983 .

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

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

[11] Dankmar Böhning,et al. Numerical estimation of a probability measure , 1985 .

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

[13] H. Wynn. The Sequential Generation of $D$-Optimum Experimental Designs , 1970 .

[14] Ilya Molchanov,et al. Variational Calculus in the Space of Measures and Optimal Design , 2001 .

[15] A. Atkinson,et al. Optimum design: 2000 , 2001 .