论文信息 - Using a Generalized Δ2-Distribution for Constructing Exact D-Optimal Designs

Using a Generalized Δ2-Distribution for Constructing Exact D-Optimal Designs

The construction of optimal designs according to a specified criterion is an optimization problem. The majority of relevant algorithms are based on generic methods in which the objective function is constructed in accordance with the chosen criterion. Focusing on the D-criterion, we will take advantage of the structure of the information matrix, whose determinant is to be maximized.

Trifon I. Missov | S. M. Ermakov | T. Missov | S. Ermakov

[1] Trifon I. Missov,et al. Integral Evaluation Using the Δ2-distribution. Simulation and Illustration , 2007, Monte Carlo Methods Appl..

[2] R. Storn,et al. Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series) , 2005 .

[3] S. M. Ermakow. Die Monte-Carlo-Methode und verwandte Fragen , 1975 .

[4] J. Tsay,et al. On the Sequential Construction of D-Optimal Designs , 1976 .

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

[6] Sergei M. Ermakov,et al. On importance sampling in the problem of global optimization , 2009, Monte Carlo Methods Appl..

[7] C. Atwood. Sequences Converging to $D$-Optimal Designs of Experiments , 1973 .

[8] W. J. Studden,et al. Optimal Experimental Designs , 1966 .

[9] O. Krafft,et al. Exact D‐Optimum Designs for Quadratic Regression , 1982 .

[10] S. M. Ermakov,et al. Polynomial Approximations and the Monte-Carlo Method , 1960 .

[11] Rainer Storn,et al. Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..