Department of Statistical Science
Duke University
presents:
David Madigan
University of Washington
"Bayesian Model Averaging"
Abstract: Standard statistical practice ignores model uncertainty. Data analysts typically select a model from some class of models and then proceed as if the model generated the data. This approach ignores a component of uncertainty, leading to over-confident inferences. Bayesian model averaging (BMA) provides a mechanism for accounting for this model uncertainty. Researchers have recently proposed several methods for implementing BMA. I will discuss these methods and present a number of applications. In these applications, BMA provides sharply improved out-of-sample predictive performance. Several of the applications use Bayesian graphical models, and the talk will describe these models in detail.
March 22, 1996
11:45 am - 12:45 pm
130 Sociology/Psychology Building Any questions concerning the seminar may be addressed to Cheryl McGhee @[919] 684-8029, e-mail cheryl@stat.duke.edu, or finger seminar@stat.duke.edu