Titre :
|
Optimal Two-Period Repeated Measurement Designs with Two or More Treatments
|
Auteurs :
|
K. Carriere ;
G. Reinsel
|
Type de document :
|
article/chapitre/communication
|
Année de publication :
|
1993
|
Format :
|
924-929
|
Langues:
|
= Anglais
|
Mots-clés:
|
Balanced
;
Crossover Design
;
Design Efficiency
;
Direct Treatment Effect
;
First-Order Residual Effect
;
Random Subject Effect
;
Repeated Measurement Design
;
Strongly Balanced
;
Universal Optimality
;
Biometrie
;
Mesures repetees
|
Résumé :
|
For two-period repeated measurement designs for comparing t greater than or equal to 2 treatments, we discuss universally optimal designs when the number of subjects N is a multiple of t(2) and some cases when N is a multiple of t. We examine efficiencies of balanced designs and other symmetric designs relative to strongly balanced designs. In addition, efficiencies of the universally optimal designs obtained by Hedayat & Zhao (1990) are examined under both fixed subject and random subject effect model assumptions.
|
Source :
|
Biometrika - 0006-3444, vol. 80, n° 4
|