Neyman & Feldman-Cousins intervals for a simple problem with an unphysical region, and an analytic solution

TL;DR

This paper compares Neyman and Feldman-Cousins confidence intervals in a simplified example, demonstrating that likelihood ratio ordering produces correct intervals while Neyman intervals can have pathologies, informing their application in particle physics measurements.

Contribution

It provides an analytic comparison of interval construction methods, illustrating the advantages of likelihood ratio ordering over traditional Neyman intervals in a specific problem.

Findings

Likelihood ratio ordering reproduces the analytic solution.

Neyman intervals can exhibit pathologies in certain cases.

Likelihood ratio intervals are preferable for the Belle phi_3 measurement.

Abstract

The new Belle phi_3/gamma measurement arXiv:hep-ex/0604054, based on Dalitz analysis of D -> Kshort pi+ pi- in B+- -> D(*) K(*)+- decays, uses likelihood ratio ordering to set confidence intervals in phi_3 and the r,delta parameters. This is different to the choice made by BaBar in PRL 95, 121802 (2005) and arXiv:hep-ex/0507101, and requires additional computation. This Note explains Belle's choice using a related but simpler example: the averaging of two numbers. We find that intervals calculated with likelihood ratio ordering reproduce the analytic solution to this problem, whereas intervals calculated by ordering according to the p.d.f. (so-called Neyman intervals) do not, and show a pathology which is important in our case. This document is adapted from a Belle Internal Note.