This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 237
Type: Contributed
Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract - #309257
Title: Modeling with Pairwise Comparisons
Author(s): Jeffrey Leek*+ and Leslie Cope and Donald Geman and Giovanni Parmigiani
Companies: Johns Hopkins Bloomberg School of Public Health and The Johns Hopkins University and The Johns Hopkins University and Harvard University
Address: , Baltimore, MD, 21401, United States
Keywords: prediction ; high-dimensional data ; pairwise comparisons ; genomics ; image analysis

We consider the problem of prediction based on a set of high-dimensional variables, or features, measured for each sample. We present a general regression framework for identifying low-dimensional predictors that are linear in the pairwise ranking of a small set of features. Our framework produces prediction functions that: (1) are rank based and robust to many aspects of measurement technology and (2) are simple to implement and interpret. For binary outcomes, our approach can be viewed as a ``linear comparison discriminant analysis'' (LCDA). We draw connections between modeling with pairwise comparisons and popular predictors like linear discriminant analysis, support vector machines, and logic regression. We illustrate our approach with examples from molecular biology, face detection, and cloud identification.

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 2010 program

2010 JSM Online Program Home

For information, contact or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.