2-TREATMENT CROSSOVER DESIGN WITH SMALL NUMBER OF PERIODS UNDER SUBJECT DROPOUT

by Shi Zhao.

Abstract: This paper focuses on two-treatment crossover designs with small number of periods, 3 or 4, assuming subjects dropout independently with a common probability from the final period. A weighted precision criterion is introduced as the design selection guidance. "Best" designs are numerically obtained under a autoregressive variance model.

Key Words: 2-treatment crossover design, Subject dropout, Precision loss, Optimality

Author:
Shi Zhao, shi.zhao@xerox.com

Editor: Johan O. Fellman, fellman@hanken.fi

READING THE ARTICLE: You can read the article in portable document (.pdf) format (191744 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 1552 times since NOVEMBER 8, 2010.


Return to the InterStat Home Page.