The gROC curve and the optimal classification
The binary classification problem (BCP) aims to correctly allocate subjects in one of two possible groups. The groups are frequently defined as having or not one characteristic of interest. With this goal, we are allowed to use different types of information. There is a huge number of methods dealin...
Uložené v:
| Vydané v: | The international journal of biostatistics |
|---|---|
| Hlavní autori: | , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Germany
03.11.2025
|
| Predmet: | |
| ISSN: | 1557-4679, 1557-4679 |
| On-line prístup: | Zistit podrobnosti o prístupe |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | The binary classification problem (BCP) aims to correctly allocate subjects in one of two possible groups. The groups are frequently defined as having or not one characteristic of interest. With this goal, we are allowed to use different types of information. There is a huge number of methods dealing with this problem; including standard binary regression models, or complex machine learning techniques such as support vector machine, boosting, or perceptron, among others. When this information is summarized in a continuous score, we have to define classification regions (or subsets) which will determine whether the subjects are classified as positive, with the characteristic under study, or as negative, otherwise. The standard (or regular) receiver-operating characteristic (ROC) curve assumes that higher values of the marker are associated with higher probabilities of being positive and considers as positive those patients with values within the intervals [
, ∞)
, and plots the true- against the false- positive rates (sensitivity against one minus specificity) for all potential
. The so-called generalized ROC curve, gROC, allows that both higher and lower values of the score are associated with higher probabilities of being positive. The efficient ROC curve, eROC, considers the best ROC curve based on a transformation of the score. In this manuscript, we are interested in studying, comparing and approximating the transformations leading to the eROC and to the gROC curves. We will prove that, when the optimal transformation does not have relative maximum, both curves are equivalent. Besides, we investigate the use of the gROC curve on some theoretical models, explore the relationship between the gROC and the eROC curves, and propose two non-parametric procedures for approximating the transformation leading to the gROC curve. The finite-sample behavior of the proposed estimators is explored through Monte Carlo simulations. Two real-data sets illustrate the practical use of the proposed methods. |
|---|---|
| AbstractList | The binary classification problem (BCP) aims to correctly allocate subjects in one of two possible groups. The groups are frequently defined as having or not one characteristic of interest. With this goal, we are allowed to use different types of information. There is a huge number of methods dealing with this problem; including standard binary regression models, or complex machine learning techniques such as support vector machine, boosting, or perceptron, among others. When this information is summarized in a continuous score, we have to define classification regions (or subsets) which will determine whether the subjects are classified as positive, with the characteristic under study, or as negative, otherwise. The standard (or regular) receiver-operating characteristic (ROC) curve assumes that higher values of the marker are associated with higher probabilities of being positive and considers as positive those patients with values within the intervals [
, ∞)
, and plots the true- against the false- positive rates (sensitivity against one minus specificity) for all potential
. The so-called generalized ROC curve, gROC, allows that both higher and lower values of the score are associated with higher probabilities of being positive. The efficient ROC curve, eROC, considers the best ROC curve based on a transformation of the score. In this manuscript, we are interested in studying, comparing and approximating the transformations leading to the eROC and to the gROC curves. We will prove that, when the optimal transformation does not have relative maximum, both curves are equivalent. Besides, we investigate the use of the gROC curve on some theoretical models, explore the relationship between the gROC and the eROC curves, and propose two non-parametric procedures for approximating the transformation leading to the gROC curve. The finite-sample behavior of the proposed estimators is explored through Monte Carlo simulations. Two real-data sets illustrate the practical use of the proposed methods. The binary classification problem (BCP) aims to correctly allocate subjects in one of two possible groups. The groups are frequently defined as having or not one characteristic of interest. With this goal, we are allowed to use different types of information. There is a huge number of methods dealing with this problem; including standard binary regression models, or complex machine learning techniques such as support vector machine, boosting, or perceptron, among others. When this information is summarized in a continuous score, we have to define classification regions (or subsets) which will determine whether the subjects are classified as positive, with the characteristic under study, or as negative, otherwise. The standard (or regular) receiver-operating characteristic (ROC) curve assumes that higher values of the marker are associated with higher probabilities of being positive and considers as positive those patients with values within the intervals [c, ∞) ( c ∈ R ) , and plots the true- against the false- positive rates (sensitivity against one minus specificity) for all potential c. The so-called generalized ROC curve, gROC, allows that both higher and lower values of the score are associated with higher probabilities of being positive. The efficient ROC curve, eROC, considers the best ROC curve based on a transformation of the score. In this manuscript, we are interested in studying, comparing and approximating the transformations leading to the eROC and to the gROC curves. We will prove that, when the optimal transformation does not have relative maximum, both curves are equivalent. Besides, we investigate the use of the gROC curve on some theoretical models, explore the relationship between the gROC and the eROC curves, and propose two non-parametric procedures for approximating the transformation leading to the gROC curve. The finite-sample behavior of the proposed estimators is explored through Monte Carlo simulations. Two real-data sets illustrate the practical use of the proposed methods.The binary classification problem (BCP) aims to correctly allocate subjects in one of two possible groups. The groups are frequently defined as having or not one characteristic of interest. With this goal, we are allowed to use different types of information. There is a huge number of methods dealing with this problem; including standard binary regression models, or complex machine learning techniques such as support vector machine, boosting, or perceptron, among others. When this information is summarized in a continuous score, we have to define classification regions (or subsets) which will determine whether the subjects are classified as positive, with the characteristic under study, or as negative, otherwise. The standard (or regular) receiver-operating characteristic (ROC) curve assumes that higher values of the marker are associated with higher probabilities of being positive and considers as positive those patients with values within the intervals [c, ∞) ( c ∈ R ) , and plots the true- against the false- positive rates (sensitivity against one minus specificity) for all potential c. The so-called generalized ROC curve, gROC, allows that both higher and lower values of the score are associated with higher probabilities of being positive. The efficient ROC curve, eROC, considers the best ROC curve based on a transformation of the score. In this manuscript, we are interested in studying, comparing and approximating the transformations leading to the eROC and to the gROC curves. We will prove that, when the optimal transformation does not have relative maximum, both curves are equivalent. Besides, we investigate the use of the gROC curve on some theoretical models, explore the relationship between the gROC and the eROC curves, and propose two non-parametric procedures for approximating the transformation leading to the gROC curve. The finite-sample behavior of the proposed estimators is explored through Monte Carlo simulations. Two real-data sets illustrate the practical use of the proposed methods. |
| Author | Martínez-Camblor, Pablo Pérez-Fernández, Sonia |
| Author_xml | – sequence: 1 givenname: Pablo orcidid: 0000-0001-7845-3905 surname: Martínez-Camblor fullname: Martínez-Camblor, Pablo organization: Faculty of Health Sciences, Universidad Autónoma de Chile, Providencia, Chile – sequence: 2 givenname: Sonia orcidid: 0000-0002-2767-6399 surname: Pérez-Fernández fullname: Pérez-Fernández, Sonia organization: Department of Statistics and Operations Research, Universidad de Oviedo, Oviedo, Asturies, Spain |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/41174954$$D View this record in MEDLINE/PubMed |
| BookMark | eNpNj8tLxDAYxIOsuA89epUevUTz5dkcpbgqLCzIei5J-lW79GXTCv73FlzB0wzDj2FmTRZt1yIh18DuQIG6r46ecsYVZQz0GVmBUoZKbezin1-SdYxHxiSkYC_IUgIYaZVcEXr4wOT9dZ8lYRq-MHFtkYxz1PVj1bg6CbWLsSqr4Maqay_JeenqiFcn3ZC37eMhe6a7_dNL9rCjgSsYaWkES6VBr4xJQ3CgHAvgQ-md06llIDwiag9CeG-thgKl1UygLl3Jmecbcvvb2w_d54RxzJsqBqxr12I3xVxwbeZbVvIZvTmhk2-wyPth3j18538X-Q9obFN6 |
| ContentType | Journal Article |
| Copyright | 2025 Walter de Gruyter GmbH, Berlin/Boston. |
| Copyright_xml | – notice: 2025 Walter de Gruyter GmbH, Berlin/Boston. |
| DBID | NPM 7X8 |
| DOI | 10.1515/ijb-2025-0016 |
| DatabaseName | PubMed MEDLINE - Academic |
| DatabaseTitle | PubMed MEDLINE - Academic |
| DatabaseTitleList | PubMed MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | no_fulltext_linktorsrc |
| Discipline | Biology |
| EISSN | 1557-4679 |
| ExternalDocumentID | 41174954 |
| Genre | Journal Article |
| GroupedDBID | --- -~S 0R~ 123 1WD 4.4 53G AAAEU AAAVF AACIX AADQG AAFPC AAFWJ AAGVJ AAILP AAJBH AALGR AAOUV AAOWA AAPJK AAQCX AARVR AASQH AAXCG ABAQN ABDRH ABFKT ABJNI ABMBZ ABMIY ABPLS ABRDF ABSOE ABWLS ABYBW ABYKJ ACDEB ACEFL ACGFO ACGFS ACHNZ ACONX ACPMA ACUND ACYCL ACZBO ADEQT ADGQD ADGYE ADJVZ ADNPR ADOZN AECWL AEDGQ AEGVQ AEICA AEJQW AEMOE AENEX AEQDQ AEQLX AERZL AFBAA AFBDD AFBQV AFCXV AFYRI AGBEV AGQYU AHCWZ AHVWV AHXUK AIWOI AKXKS ALMA_UNASSIGNED_HOLDINGS ALWYM AMVHM ASYPN BAKPI BBCWN BCIFA CFGNV CS3 DASCH DSRVY DU5 F5P HZ~ IY9 J9A K.~ KDIRW MV1 NPM NQBSW O9- P2P QD8 SA. T2Y UK5 WTRAM 7X8 |
| ID | FETCH-LOGICAL-c251t-f730847eb5778cca15a0c1bcfbaa689013beee6b133bb9961de49603e6faf20b2 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001605279100001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1557-4679 |
| IngestDate | Sat Nov 01 19:49:49 EDT 2025 Mon Nov 03 02:12:22 EST 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | eROC curve binary classification problem optimal transformation ROC curve gROC curve |
| Language | English |
| License | 2025 Walter de Gruyter GmbH, Berlin/Boston. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c251t-f730847eb5778cca15a0c1bcfbaa689013beee6b133bb9961de49603e6faf20b2 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ORCID | 0000-0002-2767-6399 0000-0001-7845-3905 |
| PMID | 41174954 |
| PQID | 3267679942 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_3267679942 pubmed_primary_41174954 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-11-03 |
| PublicationDateYYYYMMDD | 2025-11-03 |
| PublicationDate_xml | – month: 11 year: 2025 text: 2025-11-03 day: 03 |
| PublicationDecade | 2020 |
| PublicationPlace | Germany |
| PublicationPlace_xml | – name: Germany |
| PublicationTitle | The international journal of biostatistics |
| PublicationTitleAlternate | Int J Biostat |
| PublicationYear | 2025 |
| SSID | ssj0041819 |
| Score | 2.3601649 |
| SecondaryResourceType | online_first |
| Snippet | The binary classification problem (BCP) aims to correctly allocate subjects in one of two possible groups. The groups are frequently defined as having or not... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| Title | The gROC curve and the optimal classification |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/41174954 https://www.proquest.com/docview/3267679942 |
| WOSCitedRecordID | wos001605279100001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LSwMxEB7UKnjx_agvIngN3Ud2szmJFIsXaxGF3pYkm0iF7ta-wH_vJLvVkyB4yS0hzEzm-2YmzADcZHHqUNVSyzE2YYUrEgqpKUKDjVgmdZxIP2yC9_vZcCgGTcJt1nyrXPlE76iLSrsceQdpBk-5ECy6nXxQNzXKVVebERrr0IqRyjir5sPvKgJD9BK-X2rCKToE0fTYRAjvjN4VGkiUUMd5fmeXHmV6u_-93x7sNPyS3NUGsQ9rpjyArXri5OchUDQL8vb81CV6MV0aIsuCIAUkFXqOMe7Tjky730NeYUfw2rt_6T7QZmIC1chT5ijwOEC4MSrhPEPdhIkMdKi0VVKmGUJ_rIwxqcLAVCmMdMLCMAxhYpNaaaNARcewUValOQXCdJBa1w3OCKQoCOuaCZ0WPNR4ZMZtG65XcsjRIl2ZQZamWszyH0m04aQWZj6pW2fkLMQICE89-8Puc9j2GnI53PgCWhbfo7mETb2cj2bTK69qXPuDxy_CnLGh |
| linkProvider | ProQuest |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=The+gROC+curve+and+the+optimal+classification&rft.jtitle=The+international+journal+of+biostatistics&rft.au=Mart%C3%ADnez-Camblor%2C+Pablo&rft.au=P%C3%A9rez-Fern%C3%A1ndez%2C+Sonia&rft.date=2025-11-03&rft.eissn=1557-4679&rft_id=info:doi/10.1515%2Fijb-2025-0016&rft_id=info%3Apmid%2F41174954&rft_id=info%3Apmid%2F41174954&rft.externalDocID=41174954 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1557-4679&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1557-4679&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1557-4679&client=summon |