|
Activity Number:
|
113
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #304455 |
|
Title:
|
General Moments-Based Permutation Tests: A Framework, Method, and Application
|
|
Author(s):
|
Chunxiao Zhou*+ and Yongmei (Michelle) Wang
|
|
Companies:
|
University of Illinois at Urbana-Champaign and University of Illinois at Urbana-Champaign
|
|
Address:
|
115 Illini Hall , Champaign, IL, 61820,
|
|
Keywords:
|
Multiple hypotheses testing ; permutation tests ; biomedical image analysis ; moments ; partition ; recursive relation
|
|
Abstract:
|
Multiple hypotheses testing is widely used in biomedical image analysis. With unknown distribution and small sample size of the data, existing multiple testing involves testing thousands of hypotheses for statistically significant effects through permutation. The factorial scale computation cost becomes a bottle neck in the real applications. In this work, we develop a general moments-based permutation test approach to improve the system's efficiency by approximating the permutation distribution of the test statistic with Pearson distribution series, which involves the calculation of the first four moments of the permutation distribution. We thus propose a novel recursive method to derive these moments theoretically and analytically without any permutation. Detailed derivations and experimental results using different test statistics are demonstrated using simulated data and real data.
|
