|
Activity Number:
|
24
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305869 |
|
Title:
|
Factorial Crossover Designs with Fewer Periods and Fewer Subjects
|
|
Author(s):
|
Sourav Santra*+
|
|
Companies:
|
Northern Illinois University
|
|
Address:
|
Division of Statistics, DU 366, DeKalb, IL, 60115,
|
|
Keywords:
|
crossover design ; factorial experiment ; confounding
|
|
Abstract:
|
Crossover designs are commonly used in drug development. Two-period designs are the most popular because of their relative ease of implementation in practice. Factorial experiments require designs with larger number of periods. Such designs are useful as long as the number of treatment combinations, and hence the number of periods is within manageable limits from practical point of view. Another problem is that factorial crossover designs require a larger number of subjects as well. Russell and Dean (1998) constructed crossover designs with fewer subjects for two-factor experiments. Factorial crossover designs with fewer periods and fewer subjects are provided in this paper. Furthermore, designs are not restricted to two-factor experiments. An R-algorithm for constructing designs is available from the author upon request.
|
