|
Activity Number:
|
325
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #306727 |
|
Title:
|
Gaussian Process Models for a Sphere with Application to Faraday Rotation Measures
|
|
Author(s):
|
Margaret Short*+ and Dave Higdon and Philipp Kronberg
|
|
Companies:
|
Los Alamos National Laboratory and Los Alamos National Laboratory and Los Alamos National Laboratory
|
|
Address:
|
3262 Walnut Street, Apt. A, Los Alamos, NM, 87544,
|
|
Keywords:
|
Gaussian process ; spatial process ; Markov chain Monte Carlo ; Faraday rotation measures ; error mixture model
|
|
Abstract:
|
Our primary goal is to obtain a smoothed summary estimate of the magnetic field generated in and near to the Milky Way by using Faraday rotation measures (RMs). The ability to estimate the magnetic field generated locally by our galaxy and its environs will help astronomers distinguish local versus distant properties of the universe. Each RM in our dataset provides an integrated measure of the effect of the magnetic field along the entire line of sight to an extragalactic radio source. RMs can be considered prototypical of geostatistical data on a sphere. In order to model such data, we employ a Bayesian process convolution approach that uses Markov chain Monte Carlo for estimation and prediction. Complications arise due to contamination in the RM measurements, and we resolve these by means of a mixture prior on the errors. This is joint work with D. Higdon and P. Kronberg.
|
