Abstract Details
Activity Number:
|
134
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 5, 2013 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Computing
|
Abstract - #309282 |
Title:
|
Convergence Assessment in Large Finite State Space MCMC with an Application to Bayesian Clustering
|
Author(s):
|
Masoud Asgharian*+ and Ioana Ada Coxma and Vahid Partovi Nia
|
Companies:
|
McGill University and University of Ottawa and École Polytechnique Montréal
|
Keywords:
|
ayesian clustering ;
Convergence assessment ;
Gibbs sampling ;
Markov chain Monte Carlo algorithm ;
Regenerative simulation ;
Split-merge algorithm
|
Abstract:
|
An alternative to the well-known method of agglomerative Bayesian clustering is consensus clustering, in which a Markov chain Monte Carlo (MCMC) sampling algorithm explores the finite space of cluster allocations. The implementation requires diagnostic tools for assessing convergence. While there has been a surge of articles on convergence diagnostic tools for continuous stationary distributions, Bayesian clustering has spawned demands for tools designed specifically for finite state spaces. In this article we fill this gap by proposing a quantitative convergence assessment criterion for MCMC algorithms on finite state spaces that has an intuitive interpretation as a test of hypothesis. We define a one-dimensional statistic of fit and derive its asymptotic distribution. The asymptotic variance is estimated by regenerative simulation, and we perform convergence assessment via a formal statistical significance test. Finally, we consider the particular application of clustering of genetic mutants of the flowering plant {\it Arabidopsis thaliana} via metabolic profiling.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 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.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.