Was this also sent to Jugal Ghorai, Vytaras Brazauskas and Daniel Gervini (our other statisticians)? jugal@uwm.edu, vytaras@uwm.edu, gervini@uwm.edu.
Jay H. Beder, Professor EMS E485 Department of Mathematical Sciences 414-229-5280
University of Wisconsin-Milwaukee
PO Box 413
Milwaukee, WI 53201-0413
https://pantherfile.uwm.edu/beder/www/
------Original Message------
UW-Milwaukee
Lubar School of Business
Research Seminar Series
Friday April 22, 2016
10:30-12:00 Lubar Hall N440
(Refreshments at 10:30)
Izzet Sahin Memorial Lecture
Topic:
Variable selection in non-linear regression models: a parsimony-utility approach
Speaker: Robert McCulloch
Dusak Miller Professor of Econometrics and Statistics
Booth School, University of Chicago
Over the last several years, dramatic advances in Bayesian modeling and computation have given us powerful tools for flexible fitting of high dimensional relationships. However, the flexibility and complexity of the modeling procedures comes at a price: we may have difficulty understanding what our models have found. In particular, we are often interested in finding a simple model that works well, with variable selection being an important special case. Traditionally Bayesian approaches to search for a simple model have emphasized the specification of priors on models and computation of the posterior on models. In this paper we emphasize the role of utility in choosing a model. We use fits of the posterior predictive using binary tree models to search for simple structure. Tree models are computationally fast and capable of capturing complex structure so that we can feasibly search for model simplifications that are not too simple in that important variables and complexity (e.g. nonlinearity) are not lost.
About the Speaker:
Dr. McCulloch returned to Booth in 2012 from the McCombs School of Business at the University of Texas at Austin. He was previously at Booth from 1985-2008. He studies Bayesian methods, statistical analysis, machine learning and their applications in Business. He research has been published in the leading journals of statistics, econometrics, marketing, finance, and machine learning such as Statistical Science, Journal of the American Statistical Association, Biometrika, Technometrics, Journal of Econometrics, Econometric Theory, Journal of Marketing Research, Marketing Science,Review of Financial Studies, Journal of Financial Economics, and Machine Learning. Dr. McCulloch is a coauthor of Bayesian Statistics and Marketing published by Wiley in 2005. He serves at Associate Editors for the Journal of the American Statistical and Association Electronic Journal of Statistics. Professor McCulloch is a Fellow of the American Statistical Association. He holds PhD and MS in statistics from the University of Minnesota and a BS in mathematics and economics from the University of Toronto.
------------------------------
Mehdi Maadooliat
Assistant Professor, Marquette University
President, Wisconsin Chapter, ASA
------------------------------