Likelihood ratio intervals with Bayesian treatment of uncertainties: Coverage, power and combined experiments
2006 (English)In: Statistical Problems in Particle Physics, Astrophysics and Cosmology - Proceedings of PHYSTAT 2005, Imperial College Press, 2006, 93-96 p.Conference paper (Refereed)
In this note we present studies of coverage and power for confidence intervals for a Poisson process with known background calculated using the Likelihood ratio (aka Feldman k. Cousins) ordering with Bayesian treatment of uncertainties in nuisance parameters. We consider the variant where the Bayesian integration is performed in both the numerator and the denominator, and also the modification where the integration is done only in the numerator whereas in the denominator the likelihood is taken at the maximum likelihood estimate of the parameters. Furthermore we discuss how measurements can be combined in this framework and give an illustration with limits on the branching ratio of a rare B-meson decay recently presented by CDF/DO. A set of C ++ classes has been developed which can be used to calculate confidence intervals for single or combining multiple experiments using the above algorithms and considering a variety of parameterizations to describe the uncertainties.
Place, publisher, year, edition, pages
Imperial College Press, 2006. 93-96 p.
Experiments, High energy physics, Maximum likelihood estimation, B meson decays, Bayesian integration, Branching ratio, Confidence interval, Likelihood ratios, Maximum likelihood estimate, Nuisance parameter, Poisson process
IdentifiersURN: urn:nbn:se:kth:diva-164239ScopusID: 2-s2.0-84894198485ISBN: 1860946496ISBN: 9781860946493OAI: oai:DiVA.org:kth-164239DiVA: diva2:804989
5th Statistical Problems in Particle Physics, Astrophysics and Cosmology Conference, PHYSTAT 2005, 12 September 2005 through 15 September 2005, Oxford, United Kingdom
QC 201504142015-04-142015-04-142015-04-14Bibliographically approved