OPTIMAL DESIGNS FOR TWO-PARAMETER NONLINEAR MODELS WITH APPLICATION TO SURVIVAL MODELS

Censoring occurs in many industrial or biomedical ‘time to event’ experiments. Finding efficient designs for such experiments can be problematic since the statistical models involved are usually nonlinear, making the optimal choice of design parameter dependent. We provide analytical characterisatio...

Full description

Saved in:
Bibliographic Details
Published in:Statistica Sinica 2014-01, Vol.24 (1), p.415-428
Main Authors: Konstantinou, Maria, Biedermann, Stefanie, Kimber, Alan
Format: Article
Language:English
Subjects:
Censored data
Censorship
Design analysis
Experiment design
Fisher information
Mathematical vectors
Maximin
Parametric models
Regression analysis
Research universities
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Censoring occurs in many industrial or biomedical ‘time to event’ experiments. Finding efficient designs for such experiments can be problematic since the statistical models involved are usually nonlinear, making the optimal choice of design parameter dependent. We provide analytical characterisations of locallyD- andc-optimal designs for a class of models, thus reducing the numerical effort for design search substantially. We also investigate standadised maximinD- andc-optimal designs. We illustrate our results using the natural proportional hazards parameterisation of the exponential regression model. Different censoring mechanisms are incorporated and the robustness of designs against parameter misspecification is assessed.
ISSN:1017-0405
1996-8507
DOI:10.5705/ss.2011.271