dorsal/arxiv
View SchemaEvaluation of three methods for calculating statistical significance when incorporating a systematic uncertainty into a test of the background-only hypothesis for a Poisson process
| Authors | Robert D. Cousins, James T. Linnemann, Jordan Tucker |
|---|---|
| Categories | |
| ArXiv ID | physics/0702156 |
| URL | https://arxiv.org/abs/physics/0702156 |
| DOI | 10.1016/j.nima.2008.07.086 |
| Journal | Nuclear Instruments and Methods in Physics Research A 595 (2008) 480--501 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Hypothesis tests for the presence of new sources of Poisson counts amidst background processes are frequently performed in high energy physics (HEP), gamma ray astronomy (GRA), and other branches of science. While there are conceptual issues already when the mean rate of background is precisely known, the issues are even more difficult when the mean background rate has non-negligible uncertainty. After describing a variety of methods to be found in the HEP and GRA literature, we consider in detail three classes of algorithms and evaluate them over a wide range of parameter space, by the criterion of how close the ensemble-average Type I error rate (rejection of the background-only hypothesis when it is true) compares with the nominal significance level given by the algorithm. We recommend wider use of an algorithm firmly grounded in frequentist tests of the ratio of Poisson means, although for very low counts the over-coverage can be severe due to the effect of discreteness. We extend the studies of Cranmer, who found that a popular Bayesian-frequentist hybrid can undercover severely when taken to high Z values. We also examine the profile likelihood method, which has long been used in GRA and HEP; it provides an excellent approximation in much of the parameter space, as previously studied by Rolke and collaborators.
{
"annotation_id": "6b78939b-51c4-4f3d-b561-53d5678a9ba4",
"date_created": "2026-03-02T18:01:17.312000Z",
"date_modified": "2026-03-02T18:01:17.312000Z",
"file_hash": "67a112f4cba9b111f178554aaffd2e855bbb48764b376259fdef458ae8775663",
"private": false,
"record": {
"abstract": "Hypothesis tests for the presence of new sources of Poisson counts amidst\nbackground processes are frequently performed in high energy physics (HEP),\ngamma ray astronomy (GRA), and other branches of science. While there are\nconceptual issues already when the mean rate of background is precisely known,\nthe issues are even more difficult when the mean background rate has\nnon-negligible uncertainty. After describing a variety of methods to be found\nin the HEP and GRA literature, we consider in detail three classes of\nalgorithms and evaluate them over a wide range of parameter space, by the\ncriterion of how close the ensemble-average Type I error rate (rejection of the\nbackground-only hypothesis when it is true) compares with the nominal\nsignificance level given by the algorithm. We recommend wider use of an\nalgorithm firmly grounded in frequentist tests of the ratio of Poisson means,\nalthough for very low counts the over-coverage can be severe due to the effect\nof discreteness. We extend the studies of Cranmer, who found that a popular\nBayesian-frequentist hybrid can undercover severely when taken to high Z\nvalues. We also examine the profile likelihood method, which has long been used\nin GRA and HEP; it provides an excellent approximation in much of the parameter\nspace, as previously studied by Rolke and collaborators.",
"arxiv_id": "physics/0702156",
"authors": [
"Robert D. Cousins",
"James T. Linnemann",
"Jordan Tucker"
],
"categories": [
"physics.data-an"
],
"doi": "10.1016/j.nima.2008.07.086",
"journal_ref": "Nuclear Instruments and Methods in Physics Research A 595 (2008)\n 480--501",
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Evaluation of three methods for calculating statistical significance when incorporating a systematic uncertainty into a test of the background-only hypothesis for a Poisson process",
"url": "https://arxiv.org/abs/physics/0702156"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "4750d92a-cef4-43df-80d9-15a8889613af",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}