𝔖 Bobbio Scriptorium
✦   LIBER   ✦

First-order optimal designs for non-linear models

✍ Scribed by Paola Sebastiani; Raffaella Settimi

Publisher
Elsevier Science
Year
1998
Tongue
English
Weight
209 KB
Volume
74
Category
Article
ISSN
0378-3758

No coin nor oath required. For personal study only.

✦ Synopsis

This paper presents D-optimal experimental designs for a variety of non-linear models which depend on an arbitrary number of covariates but assume a positive prior mean and a Fisher information matrix satisfying particular properties. It is argued that these optimal designs can be regarded as a ΓΏrst-order approximation of the asymptotic increase of Shannon information. The e ciency of this approximation is compared in some examples, which show how the results can be further used to compute the Bayesian optimal design, when the approximate solution is not accurate enough.

πŸ“œ SIMILAR VOLUMES

G-optimal N-observation first order 2k d
G-optimal N-observation first order 2k designs
✍ S. Kounias; M. Lefkopoulou; C. Bagiatis πŸ“‚ Article πŸ“… 1983 πŸ› Elsevier Science 🌐 English βš– 381 KB

The G-optimal first order (weighing) design with k factor,s and N observations is examined. It is shown that for N =-1 (rood 4), k < N -1, the G-0ritual design is also D-optimal, whenever a Hadamard matrix of order N-I exists. The same is true for the saturated designs with N=5, 13,25 It is also sho