The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
176
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2012 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
Abstract - #306003 |
Title:
|
Tau Path Test for Subpopulation Dependency: Basics, Extensions, and Examples
|
Author(s):
|
Stephen Niklaus Bamattre*+ and Joseph Verducci
|
Companies:
|
The Ohio State University and The Ohio State University
|
Address:
|
Department of Statistics, Columbus, OH, 43210, United States
|
Keywords:
|
data mining ;
machine learning ;
nonparametrics ;
copulas ;
combinatorics
|
Abstract:
|
The tau-path is a technique to detect monotone association between a pair of variables in an unspecified population. Extracting this dependency structure can be reduced to deconvolving a mixture of two absolutely continuous distributions. Alternative approaches perform this deconvolution for monotone dependence but rely on a fixed subpopulation. The rejection bounds for detection of an associated subpopulation, previously estimated via simulation, are obtained from the underlying geometry of the association and related to the asymptotic distribution of the Longest Increasing Subsequence (LIS) of a random permutation. This allows for the procedure to be applied to large n data, where sufficient characterization of the tau-path on the basis of generated random permutations is computationally intractable. We demonstrate that this approach detects substantial relationships embedded within a marketing data set provided by Nationwide Insurance.
|
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 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.