Authors: Braess, Dietrich
Dette, Holger
Title: Optimal discriminating designs for several competing regression models
Language (ISO): en
Abstract: The problem of constructing optimal designs for a class of regression models is considered. We investigate a version of the Tp-optimality criterion as introduced by Atkinson and Fedorov (1975b) and demonstrate that optimal designs with respect to this type of criteria can be obtained by solving (nonlinear) vector-valued approximation problems. We provide a characterization of the best approximations in this context and use these results to develop an efficient algorithm for the determination of the optimal discriminating designs. The results are illustrated by fnumerical examples.
Subject Headings: model discrimination
optimal design
vector-valued approximation
URI: http://hdl.handle.net/2003/29216
http://dx.doi.org/10.17877/DE290R-3073
Issue Date: 2011-12-06
Appears in Collections:Sonderforschungsbereich (SFB) 823

Files in This Item:
File Description SizeFormat 
DP_4611_SFB823_Braess_Dette.pdfDNB190.3 kBAdobe PDFView/Open


This item is protected by original copyright



This item is protected by original copyright rightsstatements.org