|
Activity Number:
|
291
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Council of Chapters
|
| Abstract - #303062 |
|
Title:
|
The Role of Permutation and Randomization Tests in High-Dimensional Studies
|
|
Author(s):
|
Dan Nettleton*+
|
|
Companies:
|
Iowa State University
|
|
Address:
|
, , ,
|
|
Keywords:
|
false discovery rate ; microarray ; multiple testing
|
|
Abstract:
|
Permutation and randomization techniques provide reliable tests of significance when the condition of exchangeability holds under the null hypothesis. Such tests are particularly useful when parametric model assumptions are (or may be) substantially discordant with reality. Permutation-based assessments of significance were among the first methods proposed for the analysis of microarray experiments. We review some of the uses of permutation testing in high-dimensional data analyses, discuss the formal justification for such procedures, examine recent applications, and discuss strengths and weaknesses. We focus on the problem of simultaneously testing hundreds or thousands of null hypotheses and consider situations in which each null hypothesis involves the equality of either univariate or multivariate distributions.
|
- 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 |
