Self-consistent Bayesian analysis of space-time symmetry studies
Abstract
We introduce a Bayesian method for the analysis of epithermal neutron transmission data on space-time symmetries in which unique assignment of the prior is achieved by maximisation of the cross entropy and the imposition of a self-consistency criterion. Unlike the maximum likelihood method used in previous analyses of parity-violation data, our method is freed of an ad hoc cutoff parameter. Monte Carlo studies indicate that our self-consistent Bayesian analysis is superior to the maximum likelihood method when applied to the small data samples typical of symmetry studies.
- Publication:
-
Nuclear Instruments and Methods in Physics Research A
- Pub Date:
- February 1996
- DOI:
- 10.1016/0168-9002(95)00802-0
- Bibcode:
- 1996NIMPA.370..509D