JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 547
Type: Contributed
Date/Time: Wednesday, August 12, 2015 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Learning and Data Mining
Abstract #316730
Title: A Variational EM Approach for Fitting Mixed Membership Models with Rank Data
Author(s): Y. Samuel Wang* and Elena Erosheva
Companies: University of Washington and University of Washington
Keywords: Rank Data ; Variational ; Mixed Membership Models
Abstract:

Rank data occur in a wide variety of contexts including survey results, consumer preferences or athletic competitions. One way to model population heterogeneity in observed rankings is to use mixed membership models that provide a framework for individual level mixing. This framework allows each individual's membership to spread out across multiple sub-populations. Previous work on mixed membership models for rank data utilizes an MCMC approach that assumes equal-sized sub-populations. We propose a variational EM algorithm for fitting mixed membership models for rank data that allows for fast approximate inference and varying sub-population relative frequencies. We derive the variational EM algorithm, illustrate our estimation approach on data from the 1997 Irish presidential election, and compare our results to the previously published MCMC results on the same data set.


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

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home