Confidence intervals for the length of the receiver-operating characteristic curve based on a smooth estimator

A good diagnostic test should show different behavior on both the positive and the negative populations. However, this is not enough for having a good classification system. The binary classification problem is a complex task, which implies to define decision criteria. The knowledge of the level of...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Statistical methods in medical research Jg. 32; H. 5; S. 978
1. Verfasser: Martínez-Camblor, Pablo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: England 01.05.2023
Schlagworte:
ISSN:1477-0334, 1477-0334
Online-Zugang:Weitere Angaben
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract A good diagnostic test should show different behavior on both the positive and the negative populations. However, this is not enough for having a good classification system. The binary classification problem is a complex task, which implies to define decision criteria. The knowledge of the level of dissimilarity between the two involved distributions is not enough. We also have to know how to define those decision criteria. The length of the receiver-operating characteristic curve has been proposed as an index of the optimal discriminatory capacity of a biomarker. It is related not with the actual but with the optimal classification capacity of the considered diagnostic test. One particularity of this index is that its estimation should be based on parametric or smoothed models. We explore here the behavior of a kernel density estimator-based approximation for estimating the length of the receiver-operating characteristic curve. We prove the asymptotic distribution of the resulting statistic, propose a parametric bootstrap algorithm for confidence intervals construction, discuss the role that the bandwidth parameter plays in the quality of the provided estimations and, via Monte Carlo simulations, study its finite-sample behavior considering four different criteria for the bandwidth selection. The practical use of the length of the receiver-operating characteristic curve is illustrated through two real-world examples.
AbstractList A good diagnostic test should show different behavior on both the positive and the negative populations. However, this is not enough for having a good classification system. The binary classification problem is a complex task, which implies to define decision criteria. The knowledge of the level of dissimilarity between the two involved distributions is not enough. We also have to know how to define those decision criteria. The length of the receiver-operating characteristic curve has been proposed as an index of the optimal discriminatory capacity of a biomarker. It is related not with the actual but with the optimal classification capacity of the considered diagnostic test. One particularity of this index is that its estimation should be based on parametric or smoothed models. We explore here the behavior of a kernel density estimator-based approximation for estimating the length of the receiver-operating characteristic curve. We prove the asymptotic distribution of the resulting statistic, propose a parametric bootstrap algorithm for confidence intervals construction, discuss the role that the bandwidth parameter plays in the quality of the provided estimations and, via Monte Carlo simulations, study its finite-sample behavior considering four different criteria for the bandwidth selection. The practical use of the length of the receiver-operating characteristic curve is illustrated through two real-world examples.A good diagnostic test should show different behavior on both the positive and the negative populations. However, this is not enough for having a good classification system. The binary classification problem is a complex task, which implies to define decision criteria. The knowledge of the level of dissimilarity between the two involved distributions is not enough. We also have to know how to define those decision criteria. The length of the receiver-operating characteristic curve has been proposed as an index of the optimal discriminatory capacity of a biomarker. It is related not with the actual but with the optimal classification capacity of the considered diagnostic test. One particularity of this index is that its estimation should be based on parametric or smoothed models. We explore here the behavior of a kernel density estimator-based approximation for estimating the length of the receiver-operating characteristic curve. We prove the asymptotic distribution of the resulting statistic, propose a parametric bootstrap algorithm for confidence intervals construction, discuss the role that the bandwidth parameter plays in the quality of the provided estimations and, via Monte Carlo simulations, study its finite-sample behavior considering four different criteria for the bandwidth selection. The practical use of the length of the receiver-operating characteristic curve is illustrated through two real-world examples.
A good diagnostic test should show different behavior on both the positive and the negative populations. However, this is not enough for having a good classification system. The binary classification problem is a complex task, which implies to define decision criteria. The knowledge of the level of dissimilarity between the two involved distributions is not enough. We also have to know how to define those decision criteria. The length of the receiver-operating characteristic curve has been proposed as an index of the optimal discriminatory capacity of a biomarker. It is related not with the actual but with the optimal classification capacity of the considered diagnostic test. One particularity of this index is that its estimation should be based on parametric or smoothed models. We explore here the behavior of a kernel density estimator-based approximation for estimating the length of the receiver-operating characteristic curve. We prove the asymptotic distribution of the resulting statistic, propose a parametric bootstrap algorithm for confidence intervals construction, discuss the role that the bandwidth parameter plays in the quality of the provided estimations and, via Monte Carlo simulations, study its finite-sample behavior considering four different criteria for the bandwidth selection. The practical use of the length of the receiver-operating characteristic curve is illustrated through two real-world examples.
Author Martínez-Camblor, Pablo
Author_xml – sequence: 1
  givenname: Pablo
  orcidid: 0000-0001-7845-3905
  surname: Martínez-Camblor
  fullname: Martínez-Camblor, Pablo
  organization: Faculty of Health Sciences, Universidad Autonoma de Chile, Providencia, Chile
BackLink https://www.ncbi.nlm.nih.gov/pubmed/36919382$$D View this record in MEDLINE/PubMed
BookMark eNpNUMtOwzAQtFARfcAHcEE-cgn4lTg5oooCUiUucI78WLdBiV3spBJ_jwVF4rSzmtmZ1SzRzAcPCF1TckeplPekqRirCWOc0oqQkp-hBRVSFoRzMfuH52iZ0gchRBLRXKA5rxra8JotkF8H7zoL3gDu_AjxqPqEXYh43APuwe_GPQ7uZ4tgoDtCLMIBoho7v8Nmr6Iy-axLY2ewmeIRsFYJLA4eK5yGELIBZHZQY4iX6NzlALg6zRV63zy-rZ-L7evTy_phWxjOyFiUpVbOSi1VYy0oXTtDhIWy0VpKTUSlSllJEFwI7TSzSjnQpWsgN-KMlWyFbn99DzF8Tjm-HbpkoO-VhzCllslaMsqYLLP05iSd9AC2PcT8avxq_zpi34qbbiQ
ContentType Journal Article
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1177/09622802231160053
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
MEDLINE
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 Medicine
Statistics
Mathematics
EISSN 1477-0334
ExternalDocumentID 36919382
Genre Research Support, Non-U.S. Gov't
Journal Article
GroupedDBID ---
-TM
.2G
.2J
.2N
0-V
01A
0R~
123
1~K
29Q
31S
31U
31X
31Y
31Z
36B
4.4
53G
54M
5RE
5VS
6PF
7X7
88E
88I
8C1
8FE
8FG
8FI
8FJ
8R4
8R5
AABMB
AABOD
AACKU
AACMV
AACTG
AADUE
AAEJI
AAEWN
AAGGD
AAGLT
AAJIQ
AAJOX
AAJPV
AANSI
AAPEO
AAPII
AAQDB
AAQXH
AAQXI
AARDL
AARIX
AATAA
AATBZ
AAWTL
AAYTG
ABAWP
ABCCA
ABCJG
ABDLQ
ABDWY
ABEIX
ABFWQ
ABHKI
ABHQH
ABIDT
ABJCF
ABJIS
ABKRH
ABLUO
ABPGX
ABPNF
ABQKF
ABQXT
ABRHV
ABTDE
ABUJY
ABUWG
ABVFX
ABVVC
ABYTW
ACARO
ACDSZ
ACDXX
ACFEJ
ACFMA
ACGBL
ACGFS
ACGOD
ACGZU
ACIWK
ACJER
ACLHI
ACLZU
ACOFE
ACOXC
ACROE
ACRPL
ACSIQ
ACUAV
ACUIR
ACXKE
ACXMB
ADBBV
ADDLC
ADEBD
ADEIA
ADNMO
ADNON
ADRRZ
ADSTG
ADTBJ
ADUKL
ADVBO
ADYCS
AECGH
AECVZ
AEDTQ
AENEX
AEPTA
AEQLS
AERKM
AESZF
AEUHG
AEWDL
AEWHI
AEXNY
AFEET
AFKBI
AFKRA
AFKRG
AFMOU
AFQAA
AFUIA
AFWMB
AGKLV
AGNHF
AGQPQ
AGWFA
AGWNL
AHDMH
AHHFK
AHMBA
AJEFB
AJGYC
AJMMQ
AJUZI
AJVBE
AJXAJ
ALIPV
ALKWR
ALMA_UNASSIGNED_HOLDINGS
ALSLI
AMCVQ
AMVHM
ANDLU
ARALO
ARTOV
ASOEW
ASPBG
AUTPY
AUVAJ
AVWKF
AYAKG
AZFZN
AZQEC
B8O
B8R
B8Z
B93
B94
BBRGL
BDDNI
BENPR
BGLVJ
BKIIM
BPACV
BPHCQ
BSEHC
BVXVI
BYIEH
C45
CAG
CBRKF
CCPQU
CFDXU
CGR
COF
CORYS
CQQTX
CS3
CUY
CVF
DC-
DD-
DD0
DE-
DF0
DO-
DOPDO
DU5
DV7
DWQXO
D~Y
EBS
ECM
EIF
EJD
EMOBN
F5P
FEDTE
FHBDP
FYUFA
GNUQQ
GROUPED_SAGE_PREMIER_JOURNAL_COLLECTION
H13
HCIFZ
HEHIP
HF~
HMCUK
HVGLF
HZ~
J8X
K.F
K.J
L6V
M1P
M2P
M2S
M7S
N9A
NPM
O9-
OVD
P.B
P2P
PHGZM
PHGZT
PJZUB
POGQB
PPXIY
PQGLB
PQQKQ
PROAC
PRQQA
PSQYO
PTHSS
Q1R
Q2X
Q7K
Q7L
Q7X
Q82
Q83
RIG
ROL
S01
SASJQ
SAUOL
SCNPE
SDB
SFB
SFC
SFK
SFN
SFT
SGA
SGP
SGR
SGV
SGX
SGZ
SHG
SNB
SPJ
SPV
SQCSI
STM
TEORI
TN5
UKHRP
YHZ
ZONMY
ZPPRI
ZRKOI
7X8
AJHME
ID FETCH-LOGICAL-c320t-55bafd7b7a9ddeab8fc04de59bb77b046a5767e4344bfb2daafeb5f9e096fcd72
IEDL.DBID 7X8
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000949848500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1477-0334
IngestDate Sun Sep 28 16:04:04 EDT 2025
Mon Jul 21 06:06:17 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 5
Keywords binary classification problem
Asymptotic distribution
kernel density estimator
length of the curve
receiver-operating characteristic curve
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c320t-55bafd7b7a9ddeab8fc04de59bb77b046a5767e4344bfb2daafeb5f9e096fcd72
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0001-7845-3905
PMID 36919382
PQID 2787212275
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2787212275
pubmed_primary_36919382
PublicationCentury 2000
PublicationDate 2023-05-00
20230501
PublicationDateYYYYMMDD 2023-05-01
PublicationDate_xml – month: 05
  year: 2023
  text: 2023-05-00
PublicationDecade 2020
PublicationPlace England
PublicationPlace_xml – name: England
PublicationTitle Statistical methods in medical research
PublicationTitleAlternate Stat Methods Med Res
PublicationYear 2023
SSID ssj0007049
Score 2.3435972
Snippet A good diagnostic test should show different behavior on both the positive and the negative populations. However, this is not enough for having a good...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 978
SubjectTerms Algorithms
Area Under Curve
Biomarkers
Confidence Intervals
Monte Carlo Method
ROC Curve
Title Confidence intervals for the length of the receiver-operating characteristic curve based on a smooth estimator
URI https://www.ncbi.nlm.nih.gov/pubmed/36919382
https://www.proquest.com/docview/2787212275
Volume 32
WOSCitedRecordID wos000949848500001&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/eLvHCXMwpV1LS8NAEB7UitSDj_qqL1bwuphsNt3sSUQsHmzpQaG3sk_00KT29fudTVKLB0HwspDDhjD5mPlmZ_YbgFsjY2ZFaqjoeEs5M5zKCBelWcxjFqnMlFNLXkS_nw2HclAfuM3qtsqVTywdtS1MOCO_Y4gsdLNMpPeTTxqmRoXqaj1CYxMaCVKZgGoxXKuFi4r-xlwIGiUJr6uapeCS7AQhGAyOcdwJSPydYZaRprv_3288gL2aY5KHChSHsOHyFuz2vgVaZy3Y6dU19RY0A9-s5JqPIA83AKs5o-SjbIdEeBIktgQ3kzB1Zf5OCl8-oa90oauDFpOgzIwxkJgf8s_ELKZLR0KgtKTIiSKzcYHIIEHZYxyy_WN46z69Pj7TeiQDNQmL5jRNtfJWaKEk-kWlM28ibl0qtRZCY66tMH8Rjieca6-ZVco7nXrp0N7eWMFOYCsvcncGBImLiJwUjlnOU3yXcpjc-cQoJr3L4jbcrIw8QsiHOobKXbGYjdZmbsNp9adGk0qbY5R0JFLSjJ3_YfcFNMPw-Kp98RIaHi3qrmDbLNHq0-sSS7j2B70vQPnWBQ
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=Confidence+intervals+for+the+length+of+the+receiver-operating+characteristic+curve+based+on+a+smooth+estimator&rft.jtitle=Statistical+methods+in+medical+research&rft.au=Mart%C3%ADnez-Camblor%2C+Pablo&rft.date=2023-05-01&rft.issn=1477-0334&rft.eissn=1477-0334&rft.volume=32&rft.issue=5&rft.spage=978&rft_id=info:doi/10.1177%2F09622802231160053&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1477-0334&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1477-0334&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1477-0334&client=summon