Optimal Bayesian experimental design for linear models

From MaRDI portal





Optimal experimental designs are derived in this paper for estimation and prediction in Bayesian linear models. The designs derived are optimal under expected squared error loss under the assumption of normality, independence and homogeneity of variances. A geometric interpretation of the optimal designs is given. A bound is also given for the minimum number of design points. The optimality of rounding non-integer designs to integer designs is also discussed.




Cited in
(75)








This page was built for publication: Optimal Bayesian experimental design for linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q760135)