论文信息 - On Bayesian D‐optimum Design Criteria and the Equivalence Theorem in Non‐linear Models

On Bayesian D‐optimum Design Criteria and the Equivalence Theorem in Non‐linear Models

The concavity of some Bayesian D-optimality criteria is investigated and is found in some cases to depend on the prior distribution. In the case of a non-concave criterion, the standard equivalence theorem may fail, but a local version continues to apply.

John Hinde | David Firth | J. Hinde | D. Firth

[1] P. Whittle. Some General Points in the Theory of Optimal Experimental Design , 1973 .

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

[3] L. Mirsky,et al. Introduction to Linear Algebra , 1965, The Mathematical Gazette.

[4] D. Lindley. On a Measure of the Information Provided by an Experiment , 1956 .

[5] John Hinde,et al. Parameter Neutral Optimum Design for Non‐linear Models , 1997 .

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

[7] S. Lohr. Optimal Bayesian design of experiments for the one-way random effects model , 1995 .

[8] I. Olkin,et al. Inequalities: Theory of Majorization and Its Applications , 1980 .

[9] P. Laycock,et al. Optimum Experimental Designs , 1995 .

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

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

[12] W. Näther. Optimum experimental designs , 1994 .

[13] L. White. An extension of the General Equivalence Theorem to nonlinear models , 1973 .