|
Activity Number:
|
335
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #309280 |
|
Title:
|
Inference of Noncentrality Parameter of a Truncated Chi-Squared Distribution, with Application to Flexible Two-Stage Designs
|
|
Author(s):
|
Qizhai Li*+ and Kai Yu
|
|
Companies:
|
National Institutes of Health and National Institutes of Health
|
|
Address:
|
Academy of Mathematics and Systems Science, Biostatistics Branch, 6120 Executive Blvd, Rockville, MD, 20852-7244,
|
|
Keywords:
|
Flexible design ; non-centrality parameter ; double bootstrap ; non-central chi-squared distribution
|
|
Abstract:
|
We study the inference property of noncentrality parameter estimation based on samples from a truncated chi-squared distribution. This work is motivated by the flexible two-stage design using the log likelihood ratio test statistic, where the second-stage sample size is decided by the estimated noncentrality parameter of a truncated chi-squared distribution. We focus on the moment estimator due to its superior performance compared with other types of estimators, such as the MLE, and present theoretical results regarding to its existence and uniqueness. We propose a double bootstrap procedure for confidence interval constructions. Simulation studies are conducted to assess the accuracy of the moment estimation and evaluate the coverage property of the proposed confidence interval. Finally we show the application of the proposed method in the flexible two-stage design.
|
- 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 2007 program |
