Online Program Home
My Program

Abstract Details

Activity Number: 20
Type: Topic Contributed
Date/Time: Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #319158
Title: Top-K Tau-Path Subpopulation Screen for Monotone Association
Author(s): Joseph S. Verducci* and Srinath Sampath and Adriano Caloiaro and Wayne Johnson
Companies: The Ohio State University and Hamilton Capital Management and Myatt and Johnson and Myatt and Johnson
Keywords: Kendall's tau ; Generalized Mallows' model ; computational complexity ; Frank copula ; ranking ; robustness

A pair of variables that tend to rise and fall either together or in opposition are said to be monotonically associated. For certain phenomena, this tendency is causally restricted to a subpopulation, as, for example, an allergic reaction to an irritant. Previously, Yu et al. [2011] devised a method of rearranging observations to test paired data to see if such an association might be present in an unidentified subpopulation. However, the computational intensity of this seriation method limited its application to relatively small samples of data, and the test itself only judged if association were present in some subpopulation; it did not clearly identify the subsample that came from this subpopulation, especially when the whole sample tested positive. The present paper adds a "top-K" feature (Sampath and Verducci [2013]) based on a multistage ranking model that identifies a concise subsample that is likely to contain a high proportion of observations from the subpopulation in which the association is supported. Computational improvements incorporated into this top-K tau-path (TKTP) algorithm now allow the method to be extended to thousands of pairs of variables.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association