Cargando…
Confidence limits: what is the problem? Is there the solution?
This contribution to the debate on confidence limits focuses mostly on thecase of measurements with `open likelihood', in the sense that it is defined inthe text. I will show that, though a prior-free assessment of {\it confidence}is, in general, not possible, still a search result can be repor...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
CERN
2000
|
Materias: | |
Acceso en línea: | https://dx.doi.org/10.5170/CERN-2000-005.3 http://cds.cern.ch/record/427961 |
_version_ | 1780895150748205056 |
---|---|
author | D'Agostini, G. |
author_facet | D'Agostini, G. |
author_sort | D'Agostini, G. |
collection | CERN |
description | This contribution to the debate on confidence limits focuses mostly on thecase of measurements with `open likelihood', in the sense that it is defined inthe text. I will show that, though a prior-free assessment of {\it confidence}is, in general, not possible, still a search result can be reported in a mostlyunbiased and efficient way, which satisfies some desiderata which I believe areshared by the people interested in the subject. The simpler case of `closedlikelihood' will also be treated, and I will discuss why a uniform prior on asensible quantity is a very reasonable choice for most applications. In bothcases, I think that much clarity will be achieved if we remove from scientificparlance the misleading expressions `confidence intervals' and `confidence levels'. |
id | cern-427961 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2000 |
publisher | CERN |
record_format | invenio |
spelling | cern-4279612023-03-12T05:39:51Zdoi:10.5170/CERN-2000-005.3http://cds.cern.ch/record/427961engD'Agostini, G.Confidence limits: what is the problem? Is there the solution?Detectors and Experimental TechniquesThis contribution to the debate on confidence limits focuses mostly on thecase of measurements with `open likelihood', in the sense that it is defined inthe text. I will show that, though a prior-free assessment of {\it confidence}is, in general, not possible, still a search result can be reported in a mostlyunbiased and efficient way, which satisfies some desiderata which I believe areshared by the people interested in the subject. The simpler case of `closedlikelihood' will also be treated, and I will discuss why a uniform prior on asensible quantity is a very reasonable choice for most applications. In bothcases, I think that much clarity will be achieved if we remove from scientificparlance the misleading expressions `confidence intervals' and `confidence levels'.This contribution to the debate on confidence limits focuses mostly on the case of measurements with `open likelihood', in the sense that it is defined in the text. I will show that, though a prior-free assessment of {\it confidence} is, in general, not possible, still a search result can be reported in a mostly unbiased and efficient way, which satisfies some desiderata which I believe are shared by the people interested in the subject. The simpler case of `closed likelihood' will also be treated, and I will discuss why a uniform prior on a sensible quantity is a very reasonable choice for most applications. In both cases, I think that much clarity will be achieved if we remove from scientific parlance the misleading expressions `confidence intervals' and `confidence levels'.CERNhep-ex/0002055oai:cds.cern.ch:4279612000 |
spellingShingle | Detectors and Experimental Techniques D'Agostini, G. Confidence limits: what is the problem? Is there the solution? |
title | Confidence limits: what is the problem? Is there the solution? |
title_full | Confidence limits: what is the problem? Is there the solution? |
title_fullStr | Confidence limits: what is the problem? Is there the solution? |
title_full_unstemmed | Confidence limits: what is the problem? Is there the solution? |
title_short | Confidence limits: what is the problem? Is there the solution? |
title_sort | confidence limits: what is the problem? is there the solution? |
topic | Detectors and Experimental Techniques |
url | https://dx.doi.org/10.5170/CERN-2000-005.3 http://cds.cern.ch/record/427961 |
work_keys_str_mv | AT dagostinig confidencelimitswhatistheproblemistherethesolution |