Algorithms for finding locally and Bayesian optimal designs for binary dose-response models with control mortality

Smith, D.M. and Ridout, M.S. Algorithms for finding locally and Bayesian optimal designs for binary dose-response models with control mortality. Journal of Statistical Planning & Inference, 33 . pp. 463-478. (The full text of this publication is not available from this repository)

The full text of this publication is not available from this repository. (Contact us about this Publication)

Abstract

Algorithms for finding optimal designs for three-parameter binary dose–response models that incorporate control mortality are described. Locally and Bayesian optimal designs for models with a range of link functions are considered. Design criteria looked at include D-optimal, DA-optimal and V-optimal designs, together with Ds-optimal designs where the control mortality parameter is regarded as a nuisance parameter. The range of prior distributions for the Bayesian optimal designs includes uniform, trivariate normal and a combination of a bivariate normal prior for the parameters of the underlying dose–response with an independent uniform prior for the control mortality parameter.

Item Type: Article
Subjects: Q Science > QA Mathematics (inc Computing science) > QA276 Mathematical statistics
Divisions: Faculties > Science Technology and Medical Studies > School of Mathematics Statistics and Actuarial Science
Depositing User: Martin S Ridout
Date Deposited: 29 Jun 2011 15:18
Last Modified: 13 Dec 2011 13:18
Resource URI: http://kar.kent.ac.uk/id/eprint/9008 (The current URI for this page, for reference purposes)
  • Depositors only (login required):