Comment on Bayesian Analysis of Pentaquark Signals from CLAS Data, with Response to the Rep.pdfVIP

Comment on “Bayesian Analysis of Pentaquark Signals from CLAS Data”, with Response to the Reply by Ireland and Protopopsecu Robert D. Cousins Department of Physics and Astronomy, University of California, Los Angeles, CA 90095∗ (Dated: July 8, 2008) The CLAS Collaboration has published an analysis using Bayesian model selection. My Comment criticizing their use of arbitrary prior probability density functions, and a Reply by D.G. Ireland and D. Protopopsecu, have now been published as well. This paper responds to the Reply and discusses the issues in more detail, with particular emphasis on the problems of priors in Bayesian model selection. PACS numbers: 06.20.Dk, 07.05.Kf, 12.39.Mk, 14.80.-j 8 0 0 2 I. INTRODUCTION l u J The CLAS Collaboration [1], citing the pioneering work of Harold Jeffreys [2], has published a Letter [1] that claims 8 to illustrate a general method that “could be applied to any data set where a search for a new state has been carried out”, providing a “quantitative measure” for judging potential discovery results by using the formalism of Bayesian ] model selection. My Comment [3] criticizing their use of arbitrary prior probability density functions, and a Reply h p [4] by D.G. Ireland and D. Protopopsecu, have now been published as well. As I believe that the Reply did not - satisfactorily address the points raised by my Comment, and that it was not informed by other points in the articles p I cited, I elaborate in this post to th