Abstract #301371

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2002 Program page
JSM 2002 Abstract #301371
Activity Number: 368
Type: Invited
Date/Time: Thursday, August 15, 2002 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistical Graphics*
Abstract - #301371
Title: Object Oriented and Graphical Methods for Bayesian Models
Author(s): Robert McCulloch*+ and John Barnard
Affiliation(s): University of Chicago and deCode Genetics
Address: 1101 East 58th St., Chicago, Illinois, 60637, USA
Keywords: ellicitation ; utility ; decision making
Abstract:

Our goal is to take Bayesian modeling and: i.) program a model as on object using the modern object oriented programming. We believe that by thinking of a model as something that know certain things (for example, a data member representing the prior) and can do certain things (for example, draw from the prior), we can make Bayesian thinking more natural. Many decisions must be made. What model data are encapsulated? What methods must be supported in all models? How should different models interact ? And ii.) create graphical interfaces to basic model features. The goal here is hide the details of the model and associated computations and provide an interface that emphasizes real world decisions. There will be no emphasis on testing null hypotheses. Instead, the goal is to flexibly allow the user to explore model features that are relevant to utility calculations and, hence, decisions.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2002 program

JSM 2002

For information, contact meetings@amstat.org or phone (703) 684-1221.

If you have questions about the Continuing Education program, please contact the Education Department.

Revised March 2002