{"work":{"id":"c4450b72-ebcd-4cb4-948a-90d9589757fd","openalex_id":"https://openalex.org/W2019149276","doi":"10.1140/epjc/s10052-011-1554-0","arxiv_id":"1007.1727","raw_key":null,"title":"Asymptotic formulae for likelihood-based tests of new physics","authors":[{"given":"Glen","family":"Cowan","sequence":"first","affiliation":[]},{"given":"Kyle","family":"Cranmer","sequence":"additional","affiliation":[]},{"given":"Eilam","family":"Gross","sequence":"additional","affiliation":[]},{"given":"Ofer","family":"Vitells","sequence":"additional","affiliation":[]}],"authors_text":"G","year":2010,"venue":"physics.data-an","abstract":"We describe likelihood-based statistical tests for use in high energy physics for the discovery of new phenomena and for construction of confidence intervals on model parameters. We focus on the properties of the test procedures that allow one to account for systematic uncertainties. Explicit formulae for the asymptotic distributions of test statistics are derived using results of Wilks and Wald. We motivate and justify the use of a representative data set, called the \"Asimov data set\", which provides a simple method to obtain the median experimental sensitivity of a search or measurement as well as fluctuations about this expectation.","external_url":"https://arxiv.org/abs/1007.1727","cited_by_count":2550,"metadata_source":"pith","metadata_fetched_at":"2026-06-29T11:13:20.707299+00:00","pith_arxiv_id":"1007.1727","created_at":"2026-05-09T00:59:29.741061+00:00","updated_at":"2026-06-29T11:13:20.707299+00:00","title_quality_ok":true,"display_title":"Asymptotic formulae for likelihood-based tests of new physics","render_title":"Asymptotic formulae for likelihood-based tests of new physics"},"hub":{"state":{"work_id":"c4450b72-ebcd-4cb4-948a-90d9589757fd","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":101,"external_cited_by_count":2550,"distinct_field_count":9,"first_pith_cited_at":"2019-06-26T20:20:57+00:00","last_pith_cited_at":"2026-06-20T20:31:23+00:00","author_build_status":"needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-06-29T12:18:45.942270+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"method","n":30},{"context_role":"background","n":14},{"context_role":"baseline","n":1}],"polarity_counts":[{"context_polarity":"use_method","n":30},{"context_polarity":"background","n":13},{"context_polarity":"baseline","n":1},{"context_polarity":"unclear","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Asymptotic formulae for likelihood-based tests of new physics","claims":[{"claim_text":"We describe likelihood-based statistical tests for use in high energy physics for the discovery of new phenomena and for construction of confidence intervals on model parameters. We focus on the properties of the test procedures that allow one to account for systematic uncertainties. Explicit formulae for the asymptotic distributions of test statistics are derived using results of Wilks and Wald. We motivate and justify the use of a representative data set, called the \"Asimov data set\", which provides a simple method to obtain the median experimental sensitivity of a search or measurement as w","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"dijet invariant massm j1j2, and the transverse momen- tum of theb-jets,p T (b). Additional relevant observables include the missing transverse energy, /ET, the angular separation∆R(j 1, j2), the leading-lepton transverse mo- mentump T (ℓ1), and the transverse massMT (ℓ, /ET ). The signal significance is first evaluated using the Asi- mov approximation [86] in the absence of systematic un- certainty on the background, ZA = r 2 h (s+b) ln \u0010 1 + s b \u0011 −s i ,(16) wheresandbdenote the expected number","claim_type":"method","confidence":0.95,"evidence_strength":"citation_context"},{"claim_text":"doi:10.22323/1.093.0057,arXiv:1009.1003. [74] T. Junk, \"Confidence level computation for combining searches with small statistics\", Nucl. Instrum. Meth. A434(1999) 435,doi:10.1016/S0168-9002(99)00498-2, arXiv:hep-ex/9902006. [75] A. L. Read, \"Presentation of search results: theCL s technique\",J. Phys. G28(2002) 2693, doi:10.1088/0954-3899/28/10/313. [76] G. Cowan, K. Cranmer, E. Gross, and O. Vitells, \"Asymptotic formulae for likelihood-based tests of new physics\",Eur. Phys. J. C71(2011) 1554, d","claim_type":"method","confidence":0.95,"evidence_strength":"citation_context"},{"claim_text":"The general form of the Lagrangian is presented above for completeness and to highlight the potential interplay with kinetic mixing in broader contexts. For our analysis, we have implemented the scenario inFeynRules[103] and generated the cor- responding UFO model files [104, 105]. For matrix element evaluation and event generation, those are interfaced withMadGraph5_aMC@NLO[106]. On-shell decays of intermediate resonances are han- dled usingMadSpin[107], whileMadAnalysis5[108] is used for cut-b","claim_type":"method","confidence":0.95,"evidence_strength":"citation_context"},{"claim_text":"to the signal process, the uncertainty from PDF and QCD scale choices for all background processes are calculatedseparately,andfoundtobesmallerthantheothersourcesofbackgroundmodellinguncertainty. 9 Results 9.1 Statistical analysis The statistical analysis is based on the RooFit [118] and RooStats [119] frameworks to define a binned likelihood function, and the profile-likelihood ratio [120] is used as test statistic to classify the level of agreement between a model and the observed data. The an","claim_type":"method","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"polynomial to interpolate between discrete values and describe the signal scaling. Finally, one and two dimensional limits on EFT parameters are extracted by performing a Maximum Likelihood Fit in the sensitive variable, e.g.,mℓℓ in the WW channel, of the signal and background contributions to the mea- sured data. A scan of the negative log-likelihood ratio [103] for different EFT parameters results in95 % C.L. limits according to Wilks' theorem [104]. So far work was done only in the same sign ","claim_type":"method","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"defined in Section 6.2.1. In both cases, each factor𝐴 𝑗 𝑖𝑘 is defined as the fraction of generated signal events in STXS bin𝑗 that satisfy the reconstruction and event selection and populate histogram bin𝑘 of the discriminant observable in reconstructed category𝑖. 6.3 Statistical method All measurements are performed with a binned profile-likelihood-ratio fit [132]. The likelihood is built as product of Poisson terms for the fitted discriminant distributions in each differential bin or STXS cate","claim_type":"method","confidence":0.9,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks Asymptotic formulae for likelihood-based tests of new physics because it crossed a citation-hub threshold. Current citing contexts most often use it as method evidence (30 contexts).","role_counts":[{"n":30,"context_role":"method"},{"n":14,"context_role":"background"},{"n":1,"context_role":"baseline"}]},"error":null,"updated_at":"2026-06-29T09:28:38.875335+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"05c14466-2482-481e-becb-e7b978e9f5ec","orcid":null,"display_name":"Glen Cowan"},{"id":"10a7e587-f4aa-471e-b098-c910defdb6b2","orcid":null,"display_name":"Kyle Cranmer"},{"id":"88ddc000-65ec-433f-8b2e-2e606e4a7043","orcid":null,"display_name":"Eilam Gross"},{"id":"eba94ee7-8467-4c13-92ad-3c9e3cf326a2","orcid":null,"display_name":"Ofer Vitells"}]},"error":null,"updated_at":"2026-06-29T09:28:38.869119+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T17:49:31.987696+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"A New Method for Combining NLO QCD with Shower Monte Carlo Algorithms","work_id":"5ba04012-0596-4e5f-8cea-f7bd96aca319","shared_citers":11},{"title":"A general framework for implementing NLO calculations in shower Monte Carlo programs: the POWHEG BOX","work_id":"88655bbc-24f2-4be6-ab8e-45f366676ed9","shared_citers":10},{"title":"Ball, et al., Eur","work_id":"3ef74fc2-4cfa-4247-baa7-a79f7e46a1f1","shared_citers":10},{"title":"Matching NLO QCD computations with Parton Shower simulations: the POWHEG method","work_id":"7dcafae2-493f-4471-aca1-ad7115dc4ea0","shared_citers":10},{"title":"The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations","work_id":"4fdf8644-6bd0-401e-a724-5aa127e16f3d","shared_citers":10},{"title":"Alwall, R","work_id":"70c5db24-6371-4493-aa56-fa91ae131a52","shared_citers":9},{"title":"An Introduction to PYTHIA 8.2","work_id":"3215f6d0-0d43-46cf-970a-05a5894d973f","shared_citers":9},{"title":"GEANT4—a simulation toolkit","work_id":"bacf2107-98dd-46a4-b7c8-72da20595d86","shared_citers":9},{"title":"Cacciari, G.P","work_id":"e13404f0-6de9-41d3-869b-9fb9093f6c2c","shared_citers":8},{"title":"Cacciari, G.P","work_id":"85f4117a-44be-481c-bf4f-df8569c65d65","shared_citers":8},{"title":"Development of the CMS detector for the CERN LHC Run 3","work_id":"02afc3da-63cd-4f0a-be5e-39322af0a6da","shared_citers":8},{"title":"Electron and photon reconstruction and identification with the CMS experiment at the CERN LHC","work_id":"ce6f8c0c-75a4-4977-a7ff-f8b9c72c4e36","shared_citers":8},{"title":"Hayrapetyan, et al., Comput","work_id":"c1fa9a00-82a3-4101-9fd4-00680b8fbd29","shared_citers":8},{"title":"Jet energy scale and resolution in the CMS experiment in pp collisions at 8 TeV","work_id":"a8020d12-5cff-4a55-9036-5e256988af7f","shared_citers":8},{"title":"Particle-flow reconstruction and global event description with the CMS detector","work_id":"9a9aaf27-8df5-4181-adec-670c8eb9a5aa","shared_citers":8},{"title":"Performance of the CMS high-level trigger during LHC Run 2","work_id":"c3650d97-440e-40ca-a790-9c46d0e2e680","shared_citers":8},{"title":"Performance of the CMS Level-1 trigger in proton-proton collisions at √s=13 TeV","work_id":"fb4b248a-d0d5-483f-9879-9021fee6a2bf","shared_citers":8},{"title":"Performance of the CMS muon detector and muon reconstruction with proton-proton collisions at $\\sqrt{s}=$ 13 TeV","work_id":"3ffefe11-eb5a-44f5-8d9f-3176dfb92ba7","shared_citers":8},{"title":"The CMS experiment at the CERN LHC","work_id":"57f241a3-3973-4e80-a323-5716ce2d3ff1","shared_citers":8},{"title":"The CMS trigger system","work_id":"419a513c-f6c9-4ba9-94a5-97b827bea289","shared_citers":8},{"title":"Extraction and validation of a new set of CMS PYTHIA-8 tunes from underlying-event measurements","work_id":"0bfcc48b-a98b-4e04-8d65-52313e523419","shared_citers":7},{"title":"Frixione, P","work_id":"4ec32a88-75e7-4489-a2a8-d8f3a6847b16","shared_citers":7},{"title":"Junk,Confidence level computation for combining searches with small statistics, Nucl","work_id":"f3022889-85f1-486b-b843-ae0c9bd7ce2f","shared_citers":7},{"title":"Performance of missing transverse momentum reconstruction in proton-proton collisions at √s=13 TeV using the CMS detector","work_id":"e6836ef6-d5a4-4f58-acc6-1c6bce0e340f","shared_citers":7}],"time_series":[{"n":37,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T17:49:58.034003+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"items":[{"title":"Qwen3 Technical Report","outcome":"unchanged","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"counts":{"fixed":0,"merged":0,"unchanged":1,"quarantined":0,"needs_external_resolution":0},"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-14T17:49:38.385023+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Asymptotic formulae for likelihood-based tests of new physics","claims":[{"claim_text":"We describe likelihood-based statistical tests for use in high energy physics for the discovery of new phenomena and for construction of confidence intervals on model parameters. We focus on the properties of the test procedures that allow one to account for systematic uncertainties. Explicit formulae for the asymptotic distributions of test statistics are derived using results of Wilks and Wald. We motivate and justify the use of a representative data set, called the \"Asimov data set\", which provides a simple method to obtain the median experimental sensitivity of a search or measurement as w","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"dijet invariant massm j1j2, and the transverse momen- tum of theb-jets,p T (b). Additional relevant observables include the missing transverse energy, /ET, the angular separation∆R(j 1, j2), the leading-lepton transverse mo- mentump T (ℓ1), and the transverse massMT (ℓ, /ET ). The signal significance is first evaluated using the Asi- mov approximation [86] in the absence of systematic un- certainty on the background, ZA = r 2 h (s+b) ln \u0010 1 + s b \u0011 −s i ,(16) wheresandbdenote the expected number","claim_type":"method","confidence":0.95,"evidence_strength":"citation_context"},{"claim_text":"doi:10.22323/1.093.0057,arXiv:1009.1003. [74] T. Junk, \"Confidence level computation for combining searches with small statistics\", Nucl. Instrum. Meth. A434(1999) 435,doi:10.1016/S0168-9002(99)00498-2, arXiv:hep-ex/9902006. [75] A. L. Read, \"Presentation of search results: theCL s technique\",J. Phys. G28(2002) 2693, doi:10.1088/0954-3899/28/10/313. [76] G. Cowan, K. Cranmer, E. Gross, and O. Vitells, \"Asymptotic formulae for likelihood-based tests of new physics\",Eur. Phys. J. C71(2011) 1554, d","claim_type":"method","confidence":0.95,"evidence_strength":"citation_context"},{"claim_text":"The general form of the Lagrangian is presented above for completeness and to highlight the potential interplay with kinetic mixing in broader contexts. For our analysis, we have implemented the scenario inFeynRules[103] and generated the cor- responding UFO model files [104, 105]. For matrix element evaluation and event generation, those are interfaced withMadGraph5_aMC@NLO[106]. On-shell decays of intermediate resonances are han- dled usingMadSpin[107], whileMadAnalysis5[108] is used for cut-b","claim_type":"method","confidence":0.95,"evidence_strength":"citation_context"},{"claim_text":"to the signal process, the uncertainty from PDF and QCD scale choices for all background processes are calculatedseparately,andfoundtobesmallerthantheothersourcesofbackgroundmodellinguncertainty. 9 Results 9.1 Statistical analysis The statistical analysis is based on the RooFit [118] and RooStats [119] frameworks to define a binned likelihood function, and the profile-likelihood ratio [120] is used as test statistic to classify the level of agreement between a model and the observed data. The an","claim_type":"method","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"polynomial to interpolate between discrete values and describe the signal scaling. Finally, one and two dimensional limits on EFT parameters are extracted by performing a Maximum Likelihood Fit in the sensitive variable, e.g.,mℓℓ in the WW channel, of the signal and background contributions to the mea- sured data. A scan of the negative log-likelihood ratio [103] for different EFT parameters results in95 % C.L. limits according to Wilks' theorem [104]. So far work was done only in the same sign ","claim_type":"method","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"defined in Section 6.2.1. In both cases, each factor𝐴 𝑗 𝑖𝑘 is defined as the fraction of generated signal events in STXS bin𝑗 that satisfy the reconstruction and event selection and populate histogram bin𝑘 of the discriminant observable in reconstructed category𝑖. 6.3 Statistical method All measurements are performed with a binned profile-likelihood-ratio fit [132]. The likelihood is built as product of Poisson terms for the fitted discriminant distributions in each differential bin or STXS cate","claim_type":"method","confidence":0.9,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks Asymptotic formulae for likelihood-based tests of new physics because it crossed a citation-hub threshold. Current citing contexts most often use it as method evidence (30 contexts).","role_counts":[{"n":30,"context_role":"method"},{"n":14,"context_role":"background"},{"n":1,"context_role":"baseline"}]},"error":null,"updated_at":"2026-06-29T09:28:38.872895+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Asymptotic formulae for likelihood-based tests of new physics","claims":[{"claim_text":"We describe likelihood-based statistical tests for use in high energy physics for the discovery of new phenomena and for construction of confidence intervals on model parameters. We focus on the properties of the test procedures that allow one to account for systematic uncertainties. Explicit formulae for the asymptotic distributions of test statistics are derived using results of Wilks and Wald. We motivate and justify the use of a representative data set, called the \"Asimov data set\", which provides a simple method to obtain the median experimental sensitivity of a search or measurement as w","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Asymptotic formulae for likelihood-based tests of new physics because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T17:50:00.490554+00:00"}},"summary":{"title":"Asymptotic formulae for likelihood-based tests of new physics","claims":[{"claim_text":"We describe likelihood-based statistical tests for use in high energy physics for the discovery of new phenomena and for construction of confidence intervals on model parameters. We focus on the properties of the test procedures that allow one to account for systematic uncertainties. Explicit formulae for the asymptotic distributions of test statistics are derived using results of Wilks and Wald. We motivate and justify the use of a representative data set, called the \"Asimov data set\", which provides a simple method to obtain the median experimental sensitivity of a search or measurement as w","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Asymptotic formulae for likelihood-based tests of new physics because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"A New Method for Combining NLO QCD with Shower Monte Carlo Algorithms","work_id":"5ba04012-0596-4e5f-8cea-f7bd96aca319","shared_citers":11},{"title":"A general framework for implementing NLO calculations in shower Monte Carlo programs: the POWHEG BOX","work_id":"88655bbc-24f2-4be6-ab8e-45f366676ed9","shared_citers":10},{"title":"Ball, et al., Eur","work_id":"3ef74fc2-4cfa-4247-baa7-a79f7e46a1f1","shared_citers":10},{"title":"Matching NLO QCD computations with Parton Shower simulations: the POWHEG method","work_id":"7dcafae2-493f-4471-aca1-ad7115dc4ea0","shared_citers":10},{"title":"The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations","work_id":"4fdf8644-6bd0-401e-a724-5aa127e16f3d","shared_citers":10},{"title":"Alwall, R","work_id":"70c5db24-6371-4493-aa56-fa91ae131a52","shared_citers":9},{"title":"An Introduction to PYTHIA 8.2","work_id":"3215f6d0-0d43-46cf-970a-05a5894d973f","shared_citers":9},{"title":"GEANT4—a simulation toolkit","work_id":"bacf2107-98dd-46a4-b7c8-72da20595d86","shared_citers":9},{"title":"Cacciari, G.P","work_id":"e13404f0-6de9-41d3-869b-9fb9093f6c2c","shared_citers":8},{"title":"Cacciari, G.P","work_id":"85f4117a-44be-481c-bf4f-df8569c65d65","shared_citers":8},{"title":"Development of the CMS detector for the CERN LHC Run 3","work_id":"02afc3da-63cd-4f0a-be5e-39322af0a6da","shared_citers":8},{"title":"Electron and photon reconstruction and identification with the CMS experiment at the CERN LHC","work_id":"ce6f8c0c-75a4-4977-a7ff-f8b9c72c4e36","shared_citers":8},{"title":"Hayrapetyan, et al., Comput","work_id":"c1fa9a00-82a3-4101-9fd4-00680b8fbd29","shared_citers":8},{"title":"Jet energy scale and resolution in the CMS experiment in pp collisions at 8 TeV","work_id":"a8020d12-5cff-4a55-9036-5e256988af7f","shared_citers":8},{"title":"Particle-flow reconstruction and global event description with the CMS detector","work_id":"9a9aaf27-8df5-4181-adec-670c8eb9a5aa","shared_citers":8},{"title":"Performance of the CMS high-level trigger during LHC Run 2","work_id":"c3650d97-440e-40ca-a790-9c46d0e2e680","shared_citers":8},{"title":"Performance of the CMS Level-1 trigger in proton-proton collisions at √s=13 TeV","work_id":"fb4b248a-d0d5-483f-9879-9021fee6a2bf","shared_citers":8},{"title":"Performance of the CMS muon detector and muon reconstruction with proton-proton collisions at $\\sqrt{s}=$ 13 TeV","work_id":"3ffefe11-eb5a-44f5-8d9f-3176dfb92ba7","shared_citers":8},{"title":"The CMS experiment at the CERN LHC","work_id":"57f241a3-3973-4e80-a323-5716ce2d3ff1","shared_citers":8},{"title":"The CMS trigger system","work_id":"419a513c-f6c9-4ba9-94a5-97b827bea289","shared_citers":8},{"title":"Extraction and validation of a new set of CMS PYTHIA-8 tunes from underlying-event measurements","work_id":"0bfcc48b-a98b-4e04-8d65-52313e523419","shared_citers":7},{"title":"Frixione, P","work_id":"4ec32a88-75e7-4489-a2a8-d8f3a6847b16","shared_citers":7},{"title":"Junk,Confidence level computation for combining searches with small statistics, Nucl","work_id":"f3022889-85f1-486b-b843-ae0c9bd7ce2f","shared_citers":7},{"title":"Performance of missing transverse momentum reconstruction in proton-proton collisions at √s=13 TeV using the CMS detector","work_id":"e6836ef6-d5a4-4f58-acc6-1c6bce0e340f","shared_citers":7}],"time_series":[{"n":37,"year":2026}],"dependency_candidates":[]},"authors":[{"id":"88ddc000-65ec-433f-8b2e-2e606e4a7043","orcid":null,"display_name":"Eilam Gross","source":"manual","import_confidence":0.72},{"id":"05c14466-2482-481e-becb-e7b978e9f5ec","orcid":null,"display_name":"Glen Cowan","source":"manual","import_confidence":0.72},{"id":"10a7e587-f4aa-471e-b098-c910defdb6b2","orcid":null,"display_name":"Kyle Cranmer","source":"manual","import_confidence":0.72},{"id":"eba94ee7-8467-4c13-92ad-3c9e3cf326a2","orcid":null,"display_name":"Ofer Vitells","source":"manual","import_confidence":0.72}]}}