Bayesian Model Selection In Population-based Gravitational Wave Detection From Gamma-ray Bursts

Document #:

LIGO-G0900584-v1

Document type:

G - Presentations (eg Graphics)

Abstract:

A fraction of short, hard gamma-ray bursts are believed to be associated with the coalescence of binary neutron star or neutron star-black hole systems, resulting in strong emission of gravitational radiation. The cosmological distance of these sources means, however, that the gravitational wave signal is likely to be very weak by the time it reaches the Earth. Several methods, such as the binomial test and the rank-sum test have been proposed that combine gravitational wave data from a sample of gamma-ray bursts to search for the statistical signature of multiple signals. Here, we introduce a new statistical test based on the Bayesian odds ratio, where we evaluate the posterior probability that the output from multiple matched-filter searches for inspiral signals are associated with a sample of gamma-ray bursts.

Files in Document:

• with translucency (Poster_34_JClark.pdf, 673.3 kB)
• without translucency (Poster_34_JClark_v2.pdf, 673.9 kB)

Topics:

• Astrophysics / Multi-messenger

Author Groups:

• LSC

Keywords:

gamma ray burst simulated bayesian model selection