Discussion with José Bernardo on Bayesian Reference Analysis

September 27, 2011

This is the transcription of a discussion I moderated between José Bernardo, other statisticians, and a group of high energy physicists at a conference held at CERN in January 2011. The subject of the discussion was Bayesian reference analysis. Questions were asked about the use of sharp priors in Bayesian hypothesis tests, the relationship between objective Bayesian and frequentist testing, the construction, computation, and interpretation of reference posteriors, the combination of measurements, and sensitivity analysis. As the discussion unfolded, comments were made about more general issues such as the hierarchy of problem formulations, and somewhat esoteric topics such as the ratio of normal means.

This paper was published in “Proceedings of the PHYSTAT 2011 Workshop on Statistical Issues Related to Discovery Claims in Search Experiments and Unfolding,” CERN, Geneva, Switzerland, 17-20 January 2011, edited by H. B. Prosper and L. Lyons, CERN-2011-006, pp. 37-49.

It can be downloaded here.