CAST Seminar: David Dahl

Join us at the CANSSI Ontario STatistics Seminars (CAST) with

Dr. David Dahl

Professor and Chair
Department of Statistics
Brigham Young University

Free Hybrid (On Line/In Person) Event | Registration Required

Talk Title

Dependent Random Partitions by Shrinking Toward an Anchor

Abstract

Although exchangeable processes from Bayesian nonparametrics have been used as a generating mechanism for random partition models, we deviate from this paradigm to explicitly incorporate clustering information in the formulation of our random partition model. Our shrinkage partition distribution takes any partition distribution and shrinks its probability mass toward an anchor partition. We show how this provides a framework to model hierarchically-dependent and temporally-dependent random partitions. The shrinkage parameters control the degree of dependence, accommodating at its extremes both independence and complete equality. Since a priori knowledge of items may vary, our formulation allows the degree of shrinkage toward the anchor to be item-specific. Our random partition model has a tractable normalizing constant which allows for standard Markov chain Monte Carlo algorithms for posterior sampling. We prove intuitive theoretical properties for our distribution and compare it to related partition distributions. We show that our model provides better out-of-sample fit in a real data application.

Speaker Profile

David B. Dahl’s research interests are Bayesian nonparametrics, model-based clustering, random partition models, statistical computing, and bioinformatics. His work has appeared in journals including the Journal of the American Statistical Association, Annals of Applied Statistics, Journal of Computational and Graphical Statistics, and Bayesian Analysis. He teaches both undergraduate and graduate classes, including Bayesian statistics, statistical methods, and statistical computing.

David B. Dahl is Professor and Chair in the Department of Statistics at Brigham Young University. He started his academic career at Texas A&M University as an Assistant Professor in 2004, was promoted to an Associate Professor in 2010, and moved to BYU in 2012. He was promoted to Professor at BYU in 2015. He did doctoral work at the University of Wisconsin — Madison in the Department of Statistics and the Department of Biostatistics and Medical Informatics under the direction of Michael Newton, receiving his Ph.D. in 2004. He received B.S. and M.S. degrees from Brigham Young University in 1997 and 1998, working on his masters with Scott Grimshaw in the Department of Statistics. During 2007-2012, Dr. Dahl was an adjunct faculty member in the Division of Quantitative Sciences at the University of Texas, M.D. Anderson Cancer Center.


The event is finished.

Local Time

  • Timezone: America/New_York
  • Date: Nov 12 2024
  • Time: 3:30 pm - 4:30 pm
McMaster University - MDCL 3020

Location

McMaster University - MDCL 3020
Michael DeGroote Centre for Learning and Discovery, 1280 Main St W, Hamilton, ON L8S 4K1
CANSSI Ontario

Organizer

CANSSI Ontario
Email
esther.berzunza@utoronto.ca
Website
https://canssiontario.utoronto.ca

Moderator

Alexa Sochaniwsky
Alexa Sochaniwsky

PhD Student, McMaster University