Department of Statistical Science
Duke University
presents:
Dr. Kunio Tanabe
tanabe@stat.duke.edu
The Institute of Statistical Mathematics
Tokyo, Japan
"Nonparametric Empirical Bayes Method with Smoothness Priors"
Abstract: This talk is concerned with a computationally intensive empirical nonparametric Bayes method. It will be demonstrated how powerful the maximum type-II likelihood method due to Good (or Akaike's Bayesian Information Criterion) is in data analysis. It will be discussed how to incorporate inconsistent prior information safely by a data based empirical Bayes approach. Numerical treatment of improper priors and ill-conditioning of posterior likelihood in nonparametric Bayesian density estimation is also presented.
February 7, 1997
4:00 pm - 5:00 pm
116 Old Chem Building Any questions concerning the seminar may be addressed to Cheryl McGhee @ [919] 684-8029 or e-mail cheryl@stat.duke.edu. Please contact the author(s) directly for reprints etc.