Subscribe & download
Filter events
|
9:30am
to
10:30am |
|
VBI Social and Decision Analytics Seminar
(Greater Washington DC Metro Area)
Andrew J. Womack will present "Objective Bayesian Analysis in Regression Models."
Modern statistical problems involve selection (and effect estimation) of covariates from a large collection of possible predictors. This problem is compounded by biological settings where the number of covariates is often on the order of (or greater than) the number of observations. These settings require both good measures of model adequacy as well as a correction for multiple testing. In the Bayesian paradigm, this requires prior specification on the parameters of competing models and the model space itself.
In his presentation, Womack will discuss recent work on objective specification of these priors and provide consistency results for a fixed true model as the sample size increases. This work has important implications for the social, biological, and physical sciences.
Womack is a postdoctoral researcher in the Department of Statistics at the University of Florida.
|
|
