|
Activity Number:
|
429
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303647 |
|
Title:
|
Decision Analysis Using Confidence Intervals for a Binomial Parameter with Stratification
|
|
Author(s):
|
Kenneth Liu*+
|
|
Companies:
|
Merck & Co., Inc.
|
|
Address:
|
, , ,
|
|
Keywords:
|
Miettinen & Nurminen ; Type I error ; Power
|
|
Abstract:
|
Traditionally, hypothesis testing is used to accept or reject H0. Confidence intervals can also be used to make decisions about H0 while providing additional information about variability. Decision analysis using confidence intervals has been developed extensively for Gaussian data in group sequential trial designs. These methods are appropriate for binomial data when sample sizes are large. For binomial data with small sample sizes, similar theory has also been developed. However, methods do not exist for binomial data with stratification. We present a confidence interval approach to decision analysis to handle binomial data with stratification. Confidence intervals are constructed using the Miettinen & Nurminen approach. Comparisons between some existing methods are provided using simulation.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
