Sunday, September 24

MS52
Bayesian Analysis in Computational Science and Engineering

2:00 PM-4:00 PM
New Hampshire 1

Bayesian Analysis is a statistical method based on Bayes Theorem of conditional probability. It is increasingly being used in statistical estimation and uncertainty analysis, merging information from disparate sources, image analysis, geostatistics and software integration (down-scaling) and computational simulations used for decision- making. Current directions of research include application to problems of increasing complexity and size, development of Markov-Chain-Monte-Carlo computational approaches and application to ill-conditioned problems, such as image analysis. The scope of the minisymposium includes new applications of Bayesian analysis, new approaches to model and prior development, computational methods, validation and use of Bayesian analysis in decision-making.

2:00-2:25 Bayesian Analysis: Model Formulation, Validation and Scaling

Tara Athan, Organizer

2:30-2:55 On Informative Prior Models for Image Deformations

Valen E. Johnson, Duke University, USA

3:00-3:25 Confidence Intervals for Simulation-Based Predictions of Flow in Porous Media

David H. Sharp, Los Alamos National Laboratory, USA

3:30-3:55 Bayesian Estimation with a Stochastic Whale Population Simulation Model, and Problems with the Underlying Likelihood

Geof H. Givens, Colorado State University, USA.