Activity Number:
|
405
- Nonparametric Testing in Complex Data Settings
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 1, 2017 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Nonparametric Statistics
|
Abstract #323366
|
|
Title:
|
Enhancements to Graph-Theoretic Approaches to the Multivariate Two-Sample Problem
|
Author(s):
|
David Ruth* and Samuel Buttrey
|
Companies:
|
US Navy and Naval Postgraduate School
|
Keywords:
|
Multivariate statistics ;
Multivariate two-sample problem ;
Distribution-free test ;
Graph-based test ;
Nonparametric statistics ;
Classification and regression trees
|
Abstract:
|
Graph-theoretic approaches to the multivariate two-sample problem have attracted great interest in recent years. Approaches to this problem typically involve using data to construct graphs weighted by dissimilarity and then employing some function of edge counts as a test statistic to detect a group difference. We provide insights regarding how dense such graphs should be in order to maximize test power, and present enhancements to existing edge-counting approaches in order to accommodate cases such as very large sample size or samples consisting of more than two groups.
|
Authors who are presenting talks have a * after their name.