dorsal/arxiv
View SchemaA Comment on the Roe-Woodroofe Construction of Poisson Confidence Intervals
| Authors | Mark Mandelkern, Jonas Schultz |
|---|---|
| Categories | |
| ArXiv ID | physics/0004052 |
| URL | https://arxiv.org/abs/physics/0004052 |
Abstract
We consider the Roe-Woodroofe construction of confidence intervals for the case of a Poisson distributed variate where the mean is the sum of a known background and an unknown non-negative signal. We point out that the intervals do not have coverage in the usual sense but can be made to have such with a modification that does not affect the believability and other desirable features of this attractive construction. A similar modification can be used to provide coverage to the construction recently proposed by Cousins for the Gaussian-with-boundary problem.
{
"annotation_id": "0c0a9aae-2144-4d52-b8c7-368b8fb70993",
"date_created": "2026-03-02T18:00:32.487000Z",
"date_modified": "2026-03-02T18:00:32.487000Z",
"file_hash": "bbf72b8e1b6f121aa2d40820e96a95c0c20419155efd214934ca511434d38896",
"private": false,
"record": {
"abstract": "We consider the Roe-Woodroofe construction of confidence intervals for the\ncase of a Poisson distributed variate where the mean is the sum of a known\nbackground and an unknown non-negative signal. We point out that the intervals\ndo not have coverage in the usual sense but can be made to have such with a\nmodification that does not affect the believability and other desirable\nfeatures of this attractive construction. A similar modification can be used to\nprovide coverage to the construction recently proposed by Cousins for the\nGaussian-with-boundary problem.",
"arxiv_id": "physics/0004052",
"authors": [
"Mark Mandelkern",
"Jonas Schultz"
],
"categories": [
"physics.data-an",
"physics.gen-ph"
],
"title": "A Comment on the Roe-Woodroofe Construction of Poisson Confidence Intervals",
"url": "https://arxiv.org/abs/physics/0004052"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "089c8796-13d1-4ae8-a4b8-a1965be6c866",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}