University of Minnesota
Statistics
info@stat.umn.edu
612-625-8046


School of Statistics' home page.

News & Events

TED: Statistics Education

Geisser Lecture: "Bayesian Model Choice: Past, Present, Future" Merlise A Clyde, Duke University

Thurs Oct 5th at 3:30pm - Tate B-20

Abstract:

Significant advances have been made in Bayesian model selection and model averaging over the last 30 plus years in theory, computation and applications. While the Bayesian paradigm for model selection and uncertainty quantification is simple to describe, a challenge for practitioners is specification of prior distributions for the parameters defined for each candidate model. In variable selection this task becomes quickly daunting, particularly in the large p small n paradigm, as the number of models grows rapidly with the number of predictors p. Because of the difficulty of subjective prior specification, there have been a number of attempts to define conventional or objective prior distributions for Bayesian model selection ranging from Zellner’s g-prior or mixtures of g-priors to generalized ridge prior distributions. In addition to prior distributions on model specific parameters, prior probabilities on models play a key role in the large p paradigm. We discuss various criteria that have been deemed essential for model selection priors in the context of linear and generalized linear models and extensions. We highlight recent advances in theory, computation and software, and close with a discussion of challenges that need to be addressed.

Gary Oehlert returns to the School

We are pleased to have Gary back with us. Gary writes:

After ten years as an associate dean in the College of Liberal Arts, I am returning to a full-time faculty role in the School of Statistics. I spent one year as interim associate dean for faculty, 6.5 years as associate dean for planning (i.e., budgets and space, such a lovely time to deal with budgets), and 2.5 years as associate dean for undergraduate education. I learned a lot about how CLA and the University run and made a lot of contacts across many areas. I believe that I made a contribution to the success of CLA, and I am hopeful that I can use that knowledge and those contacts to contribute to the future success of the School of Statistics.

For the past ten years I have only been around Ford Hall an hour here and an hour there, teaching one course in some years. I made it to some faculty meetings but not most seminars. I did make it to one event, and when I introduced myself, one of our graduate students replied, "I thought you were a myth!" Ouch. I have missed two entire generations of graduate students, and I have faculty colleagues whom I barely know. I am glad that I served as an associate dean, but I am very happy to have decided to return to faculty status. You guys are my people, and I am looking forward to reintegrating into the life of the School. I will be spending this academic year trying to reboot my research program and leading the consulting activity. Next year will be back to the classroom.

3rd Annual Twin Cities ASA Fall Research Meeting

This is a reminder to register for the 3rd Annual Twin Cities ASA Fall Research Meeting, to be held on Monday, October 9, 2017 at the University of St. Thomas's downtown Minneapolis campus.

Registration closes in less than a week (October 3, 2017), so please register soon!

https://www.eventbrite.com/e/twin-cities-asa-chapter-2017-fall-research-conference-tickets-37831126919

Undergraduate and graduate students wishing to participate in the poster competition (with cash prizes!) are highly encouraged to register and submit poster titles and abstracts to twincitieschapterasa@gmail.com.

This year's conference will follow the pattern of previous successful years, with a morning short course titled "Multi-State Survival Models" by Terry Therneau (Mayo Clinic), afternoon invited speakers as described below, and a student poster session and competition.

* Paul Roback (St. Olaf College): Going Vague: A Collaboration with a Political Scientist and Undergraduate Statistics Researchers to Investigate Ambiguity in Congressional Candidate Issue Statements *

* * Galin Jones (University of Minnesota): Bayesian Penalized Regression * *

* * Jennifer Le-Rademacher (Mayo Clinic): Challenges in Adverse Events Attribution * * in Cancer Clinical Trials * *

* * Nate Helwig (University of Minnesota): Statistical Learning of Successful Smiles * *

* Registration is now open at the following link (which also contains additional meeting details) until October 3, 2017! *

* https://www.eventbrite.com/e/twin-cities-asa-chapter-2017-fall-research-conference-tickets-37831126919 *

* Thank you and hope to see many of you there!

Lan Wang was recently inducted as Fellow of the Institute of Mathematical Statistics

Lan Wang was recently inducted as Fellow of the Institute of Mathematical Statistics for her " ... influential contributions to high dimensional statistical theory and methodology in quantile regression and variable selection"

Congratulations, Lan!

Taylor Maki -- New Executive Office and Administrative Specialist

Taylor Maki (Mäki) is the University of Minnesota School of Statistics Executive Office and Administrative Specialist. She started on August 28th, 2017 and is the department financial processor and purchaser, key coordinator, and facilities representative. She is involved with the hiring processes of faculty and staff and serves as a friendly administrative contact in the front office. Prior to the start of the position, she performed similar duties at the STEM Education Center for three years. Taylor is a recent graduate of the University of Minnesota College of Liberal Arts and is currently pursuing a Licensure in Leadership in Health Information Technology to continue her research in health literacy. She served four years in the Pride of Minnesota Marching Band, playing tenor and alto saxophone, where she met her fiancé, Sam. Taylor enjoys alpine skiing, curling, cheering on all Minnesota sports teams, exploring her heritage, traveling, and going to the lake on the Iron Range. She has a black cat named Ramsey and a border collie named Belle.

Lexin Li elected Fellow of the ASA

Lexin Li, a 2003 graduate from the School of Statistics and currently Associate Professor at Berkeley (http://lexinli.biostat.berkeley.edu/) was recently elected Fellow of the American Statistical Association for his "... influential contributions to sufficient dimension reduction, tensor regression, and neuroimaging applications and for outstanding service to the profession"

Congratulations, Lexin!

New Assistant Professor -- Kean Ming Tan

We are please to have had Kean Ming Tan join the School this fall as a new assistant professor. Kean Ming writes,

I am thrilled to be part of the School of Statistics at University of Minnesota. I was born and raised in Penang, a beautiful island in Malaysia. I went to Purdue University for my undergraduate study. I then moved to the doctoral program in Biostatistics at University of Washington.

I am interested in developing statistical machine learning methods to uncover signal from big and complex data sets. My research covers a wide range of topic including penalized regression models, cluster analysis, classification, dimension reduction, and graphical models. I am also particularly interested in analyzing genomics and neuroscience data.

Welcome aboard, Kean Ming!

News Archive