Super-Learning of an Optimal Dynamic Treatment Rule
We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimati...
Saved in:
| Published in | The international journal of biostatistics Vol. 12; no. 1; pp. 305 - 332 |
|---|---|
| Main Authors | , |
| Format | Journal Article |
| Language | English |
| Published |
Germany
De Gruyter
01.05.2016
Walter de Gruyter GmbH |
| Subjects | |
| Online Access | Get full text |
Cover
Loading…
| Abstract | We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric, beyond possible knowledge about the treatment and censoring mechanisms. We propose data adaptive estimators of this optimal dynamic regime which are defined by sequential loss-based learning under both the blip function and weighted classification frameworks. Rather than
selecting an estimation framework and algorithm, we propose combining estimators from both frameworks using a super-learning based cross-validation selector that seeks to minimize an appropriate cross-validated risk. The resulting selector is guaranteed to asymptotically perform as well as the best convex combination of candidate algorithms in terms of loss-based dissimilarity under conditions. We offer simulation results to support our theoretical findings. |
|---|---|
| AbstractList | We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric, beyond possible knowledge about the treatment and censoring mechanisms. We propose data adaptive estimators of this optimal dynamic regime which are defined by sequential loss-based learning under both the blip function and weighted classification frameworks. Rather than a priori selecting an estimation framework and algorithm, we propose combining estimators from both frameworks using a super-learning based cross-validation selector that seeks to minimize an appropriate cross-validated risk. The resulting selector is guaranteed to asymptotically perform as well as the best convex combination of candidate algorithms in terms of loss-based dissimilarity under conditions. We offer simulation results to support our theoretical findings.We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric, beyond possible knowledge about the treatment and censoring mechanisms. We propose data adaptive estimators of this optimal dynamic regime which are defined by sequential loss-based learning under both the blip function and weighted classification frameworks. Rather than a priori selecting an estimation framework and algorithm, we propose combining estimators from both frameworks using a super-learning based cross-validation selector that seeks to minimize an appropriate cross-validated risk. The resulting selector is guaranteed to asymptotically perform as well as the best convex combination of candidate algorithms in terms of loss-based dissimilarity under conditions. We offer simulation results to support our theoretical findings. We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric, beyond possible knowledge about the treatment and censoring mechanisms. We propose data adaptive estimators of this optimal dynamic regime which are defined by sequential loss-based learning under both the blip function and weighted classification frameworks. Rather than a priori selecting an estimation framework and algorithm, we propose combining estimators from both frameworks using a super-learning based cross-validation selector that seeks to minimize an appropriate cross-validated risk. The resulting selector is guaranteed to asymptotically perform as well as the best convex combination of candidate algorithms in terms of loss-based dissimilarity under conditions. We offer simulation results to support our theoretical findings. We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric, beyond possible knowledge about the treatment and censoring mechanisms. We propose data adaptive estimators of this optimal dynamic regime which are defined by sequential loss-based learning under both the blip function and weighted classification frameworks. Rather than selecting an estimation framework and algorithm, we propose combining estimators from both frameworks using a super-learning based cross-validation selector that seeks to minimize an appropriate cross-validated risk. The resulting selector is guaranteed to asymptotically perform as well as the best convex combination of candidate algorithms in terms of loss-based dissimilarity under conditions. We offer simulation results to support our theoretical findings. We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric, beyond possible knowledge about the treatment and censoring mechanisms. We propose data adaptive estimators of this optimal dynamic regime which are defined by sequential loss-based learning under both the blip function and weighted classification frameworks. Rather than a priori selecting an estimation framework and algorithm, we propose combining estimators from both frameworks using a super-learning based cross-validation selector that seeks to minimize an appropriate cross-validated risk. The resulting selector is guaranteed to asymptotically perform as well as the best convex combination of candidate algorithms in terms of loss-based dissimilarity under conditions. We offer simulation results to support our theoretical findings. |
| Author | van der Laan, Mark J. Luedtke, Alexander R. |
| Author_xml | – sequence: 1 givenname: Alexander R. surname: Luedtke fullname: Luedtke, Alexander R. email: aluedtke@berkeley.edu organization: Division of Biostatistics, University of California, Berkeley, CA, USA – sequence: 2 givenname: Mark J. surname: van der Laan fullname: van der Laan, Mark J. email: laan@berkeley.edu organization: Division of Biostatistics, University of California, Berkeley, CA, USA |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/27227726$$D View this record in MEDLINE/PubMed |
| BookMark | eNp1kc1r3DAQxUVIaTZpj7kWQy-5qNVIlmQTKIT0ExYCSQq9CVkeb7XY0ka2U_a_r5ZNQxrakwT6vac3847JYYgBCTkF9g4kyPd-3VDOQFLGJD8gC5BS01Lp-pAsONQllVr8OCLH47hmrIQK6pfkiGvOteZqQcTNvMFEl2hT8GFVxK6wobjaTH6wffFxG-zgXXGb0E4Dhqm4nnt8RV50th_x9cN5Qr5__nR7-ZUur758u7xYUieFmKguUbaC8Q5a3bJGY9VpQNloobpOC1czJVzpyqbsHHKOWtkKpG0b3lqVKXFCPux9N3MzYOvy_8n2ZpNytrQ10Xrz90vwP80q3hvFpIJaZ4OzB4MU72YcJzP40WHf24BxHg3omgtd1Ypl9O0zdB3nFPJ4BirFKi4rXmfqzdNEj1H-7DMDdA-4FMcxYfeIADO7vkzuy-z6Mru-Mi-e8c5PdvJxN5Dv_6s636t-2X7C1OIqzdt8eZL5XzrgIJgUvwGVL6ud |
| CitedBy_id | crossref_primary_10_1111_rssa_12951 crossref_primary_10_1093_biomet_asy017 crossref_primary_10_1093_biostatistics_kxac029 crossref_primary_10_1214_19_AIHP1034 crossref_primary_10_1002_acr_24179 crossref_primary_10_1214_16_AOS1534 crossref_primary_10_1515_ijb_2020_0127 crossref_primary_10_1001_jamanetworkopen_2020_29050 crossref_primary_10_21105_joss_00512 crossref_primary_10_1186_s13063_022_06255_3 crossref_primary_10_1093_aje_kwac217 crossref_primary_10_1515_ijb_2020_0128 crossref_primary_10_1111_rssb_12483 crossref_primary_10_1177_09622802241307642 crossref_primary_10_1016_j_asoc_2021_108006 crossref_primary_10_1200_JCO_21_01957 crossref_primary_10_1214_23_EJS2157 crossref_primary_10_1515_jci_2022_0005 crossref_primary_10_1371_journal_pmed_1003020 crossref_primary_10_1146_annurev_statistics_030718_105251 crossref_primary_10_1016_j_drugalcdep_2021_109031 crossref_primary_10_1038_s41591_024_02902_1 crossref_primary_10_1093_jrsssb_qkae084 crossref_primary_10_1214_24_AOS2403 crossref_primary_10_1214_24_AOS2369 crossref_primary_10_1214_24_AOS2405 crossref_primary_10_1080_03610918_2020_1741621 crossref_primary_10_1093_aje_kwae040 crossref_primary_10_1177_0272989X241263356 crossref_primary_10_1002_sim_8822 crossref_primary_10_1001_jamanetworkopen_2019_21660 crossref_primary_10_1093_aje_kwae122 crossref_primary_10_1146_annurev_clinpsy_050817_084746 crossref_primary_10_1007_s10985_022_09566_4 crossref_primary_10_1111_add_15654 crossref_primary_10_1177_2167702618815466 crossref_primary_10_21105_joss_02232 crossref_primary_10_1080_01621459_2020_1828091 crossref_primary_10_1109_ACCESS_2022_3160179 crossref_primary_10_1287_msom_2023_1196 crossref_primary_10_1080_19345747_2023_2290546 crossref_primary_10_1080_01621459_2019_1668794 crossref_primary_10_1038_s41586_023_06501_x crossref_primary_10_2139_ssrn_3982796 crossref_primary_10_1007_s00127_024_02694_2 crossref_primary_10_6339_24_JDS1143 crossref_primary_10_1186_s13063_022_06438_y crossref_primary_10_1080_01621459_2016_1242427 crossref_primary_10_1080_10503307_2023_2182241 crossref_primary_10_1093_jrsssb_qkad017 crossref_primary_10_1093_biomet_asaa076 crossref_primary_10_1016_j_cct_2023_107306 crossref_primary_10_1093_jrsssb_qkac001 crossref_primary_10_1177_09622802231181223 crossref_primary_10_1093_jrsssa_qnae089 crossref_primary_10_1093_biomet_asad042 crossref_primary_10_1177_09622802211052831 crossref_primary_10_1186_s12888_022_04478_x crossref_primary_10_1080_02664763_2018_1441383 crossref_primary_10_1001_jamapsychiatry_2021_2500 crossref_primary_10_1080_07350015_2020_1811102 crossref_primary_10_1016_j_artmed_2020_101964 crossref_primary_10_1016_j_arcontrol_2018_05_001 crossref_primary_10_1017_S0033291721000027 crossref_primary_10_1080_19466315_2019_1647874 |
| Cites_doi | 10.1007/s12561-011-9038-1 10.1007/978-1-4419-9076-1_11 10.1097/QAD.0b013e32830f97e2 10.1111/1467-9868.00389 10.1080/01621459.2012.695674 10.1198/016214505000000907 10.1214/10-AOS864 10.1093/aje/kwm232 10.1080/01621459.1986.10478354 10.1037/h0037350 10.1198/jasa.2009.0011 10.1023/A:1017928328829 10.1002/sim.3301 |
| ContentType | Journal Article |
| Copyright | Copyright Walter de Gruyter GmbH 2016 |
| Copyright_xml | – notice: Copyright Walter de Gruyter GmbH 2016 |
| DBID | AAYXX CITATION CGR CUY CVF ECM EIF NPM 3V. 7X7 7XB 88C 88E 8FI 8FJ 8FK ABUWG AFKRA BENPR CCPQU FYUFA GHDGH K9. M0S M0T M1P PHGZM PHGZT PJZUB PKEHL PPXIY PQEST PQQKQ PQUKI PRINS 7X8 5PM |
| DOI | 10.1515/ijb-2015-0052 |
| DatabaseName | CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed ProQuest Central (Corporate) Health & Medical Collection ProQuest Central (purchase pre-March 2016) Healthcare Administration Database (Alumni) Medical Database (Alumni Edition) Hospital Premium Collection Hospital Premium Collection (Alumni Edition) ProQuest Central (Alumni) (purchase pre-March 2016) ProQuest Central (Alumni) ProQuest Central UK/Ireland ProQuest Central ProQuest One Community College Health Research Premium Collection Health Research Premium Collection (Alumni) ProQuest Health & Medical Complete (Alumni) ProQuest Health & Medical Collection Healthcare Administration Database Medical Database ProQuest Central Premium ProQuest One Academic (New) ProQuest Health & Medical Research Collection ProQuest One Academic Middle East (New) ProQuest One Health & Nursing ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China MEDLINE - Academic PubMed Central (Full Participant titles) |
| DatabaseTitle | CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) ProQuest One Academic Middle East (New) ProQuest One Academic Eastern Edition ProQuest Health & Medical Complete (Alumni) ProQuest Health Management ProQuest Central (Alumni Edition) ProQuest One Community College ProQuest One Health & Nursing ProQuest Hospital Collection Health Research Premium Collection (Alumni) ProQuest Central China ProQuest Hospital Collection (Alumni) ProQuest Central ProQuest Health & Medical Complete Health Research Premium Collection ProQuest Medical Library ProQuest One Academic UKI Edition Health and Medicine Complete (Alumni Edition) Health & Medical Research Collection ProQuest Health Management (Alumni Edition) ProQuest Central (New) ProQuest One Academic ProQuest One Academic (New) ProQuest Medical Library (Alumni) ProQuest Central (Alumni) MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic MEDLINE CrossRef ProQuest One Academic Middle East (New) |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: EIF name: MEDLINE url: https://proxy.k.utb.cz/login?url=https://www.webofscience.com/wos/medline/basic-search sourceTypes: Index Database – sequence: 3 dbid: BENPR name: ProQuest Central url: https://www.proquest.com/central sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Biology |
| EISSN | 1557-4679 |
| EndPage | 332 |
| ExternalDocumentID | PMC6056197 4306417471 27227726 10_1515_ijb_2015_0052 10_1515_ijb_2015_0052121305 |
| Genre | Research Support, U.S. Gov't, Non-P.H.S Journal Article Research Support, N.I.H., Extramural |
| GrantInformation_xml | – fundername: NIAID NIH HHS grantid: R01 AI074345 |
| GroupedDBID | --- -~S 0R~ 123 1WD 4.4 53G 7X7 88E 8FI 8FJ AAAEU AAAVF AACIX AADQG AAFPC AAGVJ AAILP AAJBH AALGR AAONY AAOUV AAOWA AAPJK AAQCX AARVR AASQH AAXCG ABAQN ABDRH ABFKT ABJNI ABMBZ ABMIY ABPLS ABRDF ABRQL ABSOE ABUWG ABWLS ABYBW ABYKJ ACDEB ACEFL ACGFO ACGFS ACHNZ ACMKP ACONX ACPMA ACUND ACXLN ACYCL ACZBO ADBBV ADEQT ADGQD ADGYE ADJVZ ADOZN AECWL AEDGQ AEGVQ AEICA AEJQW AEKEB AEMOE AENEX AEQDQ AEQLX AERZL AFBAA AFBDD AFBQV AFCXV AFGNR AFKRA AFYRI AGBEV AGQYU AHCWZ AHVWV AHXUK AIKXB AIWOI AKXKS ALMA_UNASSIGNED_HOLDINGS ALUKF ALWYM AMAVY AMVHM AQUVI ASYPN AZMOX BAKPI BBCWN BCIFA BENPR BLHJL BPHCQ BVXVI CCPQU CFGNV CKPZI CS3 DASCH DU5 EBS EJD F5P FYUFA HMCUK HZ~ IY9 J9A K.~ KDIRW LG7 M0T M1P MV1 NQBSW O9- P2P PHGZM PHGZT PPXIY PQQKQ PROAC PSQYO QD8 SA. T2Y UK5 UKHRP WTRAM AAYXX CITATION CGR CUY CVF ECM EIF NPM PJZUB 3V. 7XB 8FK K9. PKEHL PQEST PQUKI PRINS 7X8 5PM ADNPR |
| ID | FETCH-LOGICAL-c533t-74e5d302f1d7d0b7e8f71e5b736ff73c9063c4c4b4fce22e76a815adb2da6b733 |
| IEDL.DBID | 7X7 |
| ISSN | 2194-573X 1557-4679 |
| IngestDate | Thu Aug 21 18:42:16 EDT 2025 Fri Jul 11 02:28:25 EDT 2025 Fri Jul 25 05:45:53 EDT 2025 Mon Jul 21 05:59:47 EDT 2025 Thu Apr 24 22:59:14 EDT 2025 Tue Jul 01 01:02:50 EDT 2025 Thu Jul 10 10:39:49 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| Language | English |
| License | This content is free. http://creativecommons.org/licenses/by-nc-nd/3.0 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c533t-74e5d302f1d7d0b7e8f71e5b736ff73c9063c4c4b4fce22e76a815adb2da6b733 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| OpenAccessLink | https://proxy.k.utb.cz/login?url=https://www.degruyter.com/document/doi/10.1515/ijb-2015-0052/html |
| PMID | 27227726 |
| PQID | 1860825829 |
| PQPubID | 2031306 |
| PageCount | 28 |
| ParticipantIDs | pubmedcentral_primary_oai_pubmedcentral_nih_gov_6056197 proquest_miscellaneous_1792378960 proquest_journals_1860825829 pubmed_primary_27227726 crossref_primary_10_1515_ijb_2015_0052 crossref_citationtrail_10_1515_ijb_2015_0052 walterdegruyter_journals_10_1515_ijb_2015_0052121305 |
| ProviderPackageCode | CITATION AAYXX |
| PublicationCentury | 2000 |
| PublicationDate | 2016-05-01 |
| PublicationDateYYYYMMDD | 2016-05-01 |
| PublicationDate_xml | – month: 05 year: 2016 text: 2016-05-01 day: 01 |
| PublicationDecade | 2010 |
| PublicationPlace | Germany |
| PublicationPlace_xml | – name: Germany – name: Berlin |
| PublicationTitle | The international journal of biostatistics |
| PublicationTitleAlternate | Int J Biostat |
| PublicationYear | 2016 |
| Publisher | De Gruyter Walter de Gruyter GmbH |
| Publisher_xml | – name: De Gruyter – name: Walter de Gruyter GmbH |
| References | Laan (ref321) 2014 (ref711) 2011 (ref721) 2011 (ref581) 2003; 65 Diaz (ref381) 2013 (ref701) 2014 Dudoit (ref211) 2005; 22 (ref281) 2012; 68 ref11 Laan (ref491) 2004 (ref231) 2002; 49 (ref311) 2011 (ref261) 2012; 107 (ref361) 2006; 101 (ref301) 2011 Laan (ref731) 2014 (ref181) 2003; 65 (ref341) 2001; 29 (ref131) 2008; 27 (ref671) 2012; 107 (ref101) 2005; 10 (ref441) 1986; 81 (ref471) 1974; 66 Ypma (ref401) 2014 (ref411) 1987; 14 (ref651) 2003 Polley (ref801) 2012 (ref171) 2003; 65 Laan (ref81) 2004 (ref751) 2001; 29 (ref191) 1998 (ref201) 2004; volume 179 (ref61) 1974; 66 (ref241) 2003 (ref31) 1986; 81 (ref611) 2004; volume 179 (ref661) 2011; 39 (ref41) 1923; 5 (ref331) 2014b; 3 (ref01) 1987; 14 (ref111) 2011; 3 (ref51) 2009 (ref291) 2014 (ref741) 2014b; 3 (ref551) 2008; 22 ref421 Chakraborty (ref351) 2013 (ref221) 2005; 6 (ref531) 2010; 6 (ref521) 2011; 3 (ref631) 2005; 6 (ref501) 2007; 6 Zheng (ref781) 2010 (ref451) 1923; 5 (ref431) 1997 (ref161) 2009; 104 (ref601) 1998 (ref691) 2012; 68 (ref251) 2011; 39 (ref461) 2009 (ref151) 2007; 166 (ref561) 2007; 166 Chakraborty (ref761) 2013 (ref641) 2002; 49 (ref271) 2012; 8 (ref91) 2007; 6 (ref141) 2008; 22 (ref121) 2010; 6 (ref681) 2012; 8 (ref591) 2003; 65 Zheng (ref371) 2010 (ref21) 1997 (ref571) 2009; 104 (ref511) 2005; 10 Polley (ref391) 2012 (ref541) 2008; 27 Dudoit (ref621) 2005; 22 (ref771) 2006; 101 Ypma (ref811) 2014 Diaz (ref791) 2013 |
| References_xml | – volume: 65 start-page: 355 year: 2003 ident: ref181 article-title: Discussion of “Optimal dynamic treatment regimes” by Susan A. Murphy publication-title: J R Stat Soc Ser B – volume-title: Reinforcement learning: an introduction. year: 1998 ident: ref191 – start-page: 39 year: 2013 ident: ref791 article-title: van der Targeted data adaptive estimation of the causal dose response curve Technical Report Division of Biostatistics University of California Berkeley submitted to JCI – volume: 3 start-page: 28 year: 2011 ident: ref111 article-title: A data augmentation method for estimating the causal effect of adherence to treatment regimens targeting control of an intermediate measure publication-title: Stat Biosc doi: 10.1007/s12561-011-9038-1 – volume-title: Unified methods for censored longitudinal data and causality. year: 2003 ident: ref651 – volume: volume 179, start-page: 189 year: 2004 ident: ref611 article-title: Optimal structural nested models for optimal sequential decisions publication-title: Proc Second Seattle Symp Biostat doi: 10.1007/978-1-4419-9076-1_11 – volume: 22 start-page: 2097 year: 2008 ident: ref141 article-title: Long-term consequences of the delay between virologic failure of highly active antiretroviral therapy and regimen modification publication-title: AIDS doi: 10.1097/QAD.0b013e32830f97e2 – year: 2014 ident: ref701 article-title: New statistical learning methods for estimating optimal dynamic treatment regimes publication-title: J Am Stat Assoc – start-page: 329 year: 2014 ident: ref321 article-title: van der AR Targeted learning of an optimal dynamic treatment and statistical inference for its mean outcome Technical Report available at http www bepress com ucbbiostat Division of Biostatistics University of California a – volume: 22 start-page: 2097 year: 2008 ident: ref551 article-title: Long-term consequences of the delay between virologic failure of highly active antiretroviral therapy and regimen modification publication-title: AIDS doi: 10.1097/QAD.0b013e32830f97e2 – volume: 65 start-page: 331 year: 2003 ident: ref581 article-title: Optimal dynamic treatment regimes publication-title: J R Stat Soc Ser B doi: 10.1111/1467-9868.00389 – volume: 8 start-page: Article 18 year: 2012 ident: ref681 article-title: Statistical issues and limitations in personalized medicine research with clinical trials publication-title: Int J Biostat – volume: 22 start-page: 131 year: 2005 ident: ref621 article-title: der Asymptotics of cross - validated risk estimation in estimator selection and performance assessment publication-title: Stat – start-page: 273 year: 2010 ident: ref371 article-title: van der Asymptotic theory for cross - validated targeted maximum likelihood estimation Technical Report Division of University of California publication-title: Biostatistics – volume: 107 start-page: 1106 year: 2012 ident: ref261 article-title: Estimating individual treatment rules using outcome weighted learning publication-title: J Am Stat Assoc doi: 10.1080/01621459.2012.695674 – volume: 101 start-page: 138 year: 2006 ident: ref771 article-title: Convexity, classification, and risk bounds publication-title: J Am Stat Assoc doi: 10.1198/016214505000000907 – volume: 6 start-page: Article 25 year: 2007 ident: ref91 article-title: Super learner publication-title: Stat Appl Genet Mol – volume: 39 start-page: 1180 year: 2011 ident: ref251 article-title: Performance guarantees for individualized treatment rules publication-title: Ann Statist doi: 10.1214/10-AOS864 – volume: 5 start-page: 465 year: 1923 ident: ref41 article-title: Sur les applications de la theorie des probabilites aux experiences agricoles: Essai des principes (In Polish). English translation by D.M. Dabrowska and T.P. Speed (1990) publication-title: Stat Sci – volume: 5 start-page: 465 year: 1923 ident: ref451 article-title: Sur les applications de la theorie des probabilites aux experiences agricoles: Essai des principes (In Polish). English translation by D.M. Dabrowska and T.P. Speed (1990) publication-title: Stat Sci – volume: 166 start-page: 985 year: 2007 ident: ref151 article-title: History-adjusted marginal structural models for estimating time-varying effect modification publication-title: Am J Epidemiol doi: 10.1093/aje/kwm232 – year: 2013 ident: ref761 article-title: Statistical for Dynamic Treatment Regimes Heidelberg New York publication-title: Methods – start-page: 329 year: 2014 ident: ref731 article-title: van der AR Targeted learning of an optimal dynamic treatment and statistical inference for its mean outcome Technical Report available at http www bepress com ucbbiostat Division of Biostatistics University of California a – start-page: 142 year: 2004 ident: ref81 article-title: van der van der The cross validated adaptive epsilon net estimator Technical Report Division of University of California publication-title: Biostatistics – year: 2012 ident: ref391 article-title: der Super Learner Prediction – volume-title: Targeted Learning: Causal Inference for Observational and Experimental Data. year: 2011 ident: ref711 – volume-title: Targeted Learning: Causal Inference for Observational and Experimental Data. year: 2011 ident: ref301 – volume: 68 start-page: 103 year: 2012 ident: ref691 article-title: Estimating optimal treatment regimes from a classification perspective publication-title: Statistics – volume: 22 start-page: 131 year: 2005 ident: ref211 article-title: der Asymptotics of cross - validated risk estimation in estimator selection and performance assessment publication-title: Stat – volume: 6 start-page: 503 year: 2005 ident: ref221 article-title: Tree-based batch mode reinforcement learning publication-title: J Mach Learn Res – volume: 6 start-page: Article 25 year: 2007 ident: ref501 article-title: Super learner publication-title: Stat Appl Genet Mol – volume-title: Reinforcement learning: an introduction. year: 1998 ident: ref601 – volume: 81 start-page: 945 year: 1986 ident: ref31 article-title: Statistics and causal inference publication-title: J Am Stat Assoc doi: 10.1080/01621459.1986.10478354 – volume: 66 start-page: 688 year: 1974 ident: ref471 article-title: Estimating causal effects of treatments in randomized and nonrandomized studies publication-title: J Educ Psychol doi: 10.1037/h0037350 – volume: 6 start-page: 503 year: 2005 ident: ref631 article-title: Tree-based batch mode reinforcement learning publication-title: J Mach Learn Res – volume: 6 start-page: 8 year: 2010 ident: ref531 article-title: Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, part I: proofs and additional results publication-title: Int J Biostat – ident: ref421 article-title: Information recovery and bias adjustment in proportional hazards regression analysis of randomized trials using surrogate markers publication-title: Proc Biopharm Sect – start-page: 142 year: 2004 ident: ref491 article-title: van der van der The cross validated adaptive epsilon net estimator Technical Report Division of University of California publication-title: Biostatistics – volume: 29 start-page: 1785 year: 2001 ident: ref751 article-title: Causal inference in complex longitudinal studies: continuous case publication-title: Ann Stat – volume: 104 start-page: 155 year: 2009 ident: ref571 article-title: Estimating response-maximized decision rules with applications to breastfeeding publication-title: J Am Stat Assoc doi: 10.1198/jasa.2009.0011 – volume-title: Targeted learning: causal inference for observational and experimental data. year: 2011 ident: ref311 – year: 2014 ident: ref811 article-title: The nonlinear optimization package – start-page: 273 year: 2010 ident: ref781 article-title: van der Asymptotic theory for cross - validated targeted maximum likelihood estimation Technical Report Division of University of California publication-title: Biostatistics – volume: 14 start-page: 139s year: 1987 ident: ref411 article-title: A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods publication-title: Comput Math Appl – volume: 107 start-page: 1106 year: 2012 ident: ref671 article-title: Estimating individual treatment rules using outcome weighted learning publication-title: J Am Stat Assoc doi: 10.1080/01621459.2012.695674 – volume: 29 start-page: 1785 year: 2001 ident: ref341 article-title: Causal inference in complex longitudinal studies: continuous case publication-title: Ann Stat – volume: 49 start-page: 161 year: 2002 ident: ref231 article-title: Kernel-based reinforcement learning publication-title: Mach Learn doi: 10.1023/A:1017928328829 – year: 2014 ident: ref291 article-title: New statistical learning methods for estimating optimal dynamic treatment regimes publication-title: J Am Stat Assoc – volume: 104 start-page: 155 year: 2009 ident: ref161 article-title: Estimating response-maximized decision rules with applications to breastfeeding publication-title: J Am Stat Assoc doi: 10.1198/jasa.2009.0011 – year: 2013 ident: ref351 article-title: Statistical for Dynamic Treatment Regimes Heidelberg New York publication-title: Methods – year: 2014 ident: ref401 article-title: The nonlinear optimization package – volume: 81 start-page: 945 year: 1986 ident: ref441 article-title: Statistics and causal inference publication-title: J Am Stat Assoc doi: 10.1080/01621459.1986.10478354 – volume: 27 start-page: 4678 year: 2008 ident: ref131 article-title: Estimation and extrapolation of optimal treatment and testing strategies publication-title: Stat Med doi: 10.1002/sim.3301 – volume: 27 start-page: 4678 year: 2008 ident: ref541 article-title: Estimation and extrapolation of optimal treatment and testing strategies publication-title: Stat Med doi: 10.1002/sim.3301 – volume: 39 start-page: 1180 year: 2011 ident: ref661 article-title: Performance guarantees for individualized treatment rules publication-title: Ann Statist doi: 10.1214/10-AOS864 – volume: 3 start-page: 61 year: 2014b ident: ref741 article-title: Targeted learning of the mean outcome under an optimal dynamic treatment rule publication-title: J Causal Inference – volume: 49 start-page: 161 year: 2002 ident: ref641 article-title: Kernel-based reinforcement learning publication-title: Mach Learn doi: 10.1023/A:1017928328829 – volume: 3 start-page: 61 year: 2014b ident: ref331 article-title: Targeted learning of the mean outcome under an optimal dynamic treatment rule publication-title: J Causal Inference – volume-title: Unified methods for censored longitudinal data and causality. year: 2003 ident: ref241 – volume: 6 start-page: 8 year: 2010 ident: ref121 article-title: Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, part I: proofs and additional results publication-title: Int J Biostat – volume: 65 start-page: 355 year: 2003 ident: ref591 article-title: Discussion of “Optimal dynamic treatment regimes” by Susan A. Murphy publication-title: J R Stat Soc Ser B – volume: 68 start-page: 103 year: 2012 ident: ref281 article-title: Estimating optimal treatment regimes from a classification perspective publication-title: Statistics – volume: 3 start-page: 28 year: 2011 ident: ref521 article-title: A data augmentation method for estimating the causal effect of adherence to treatment regimens targeting control of an intermediate measure publication-title: Stat Biosc doi: 10.1007/s12561-011-9038-1 – volume: 8 start-page: Article 18 year: 2012 ident: ref271 article-title: Statistical issues and limitations in personalized medicine research with clinical trials publication-title: Int J Biostat – volume-title: Causality: models, reasoning and inference, year: 2009 ident: ref51 – volume: 10 start-page: 24 year: 2005 ident: ref101 article-title: An experimental design for the development of adaptive treatment strategies publication-title: Stat Med – volume-title: Causality: models, reasoning and inference, year: 2009 ident: ref461 – volume: 14 start-page: 139s year: 1987 ident: ref01 article-title: A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods publication-title: Comput Math Appl – volume: 101 start-page: 138 year: 2006 ident: ref361 article-title: Convexity, classification, and risk bounds publication-title: J Am Stat Assoc doi: 10.1198/016214505000000907 – year: 2012 ident: ref801 article-title: der Super Learner Prediction – start-page: 69 volume-title: Latent variable modeling and applications to causality year: 1997: ident: ref21 – volume: 166 start-page: 985 year: 2007 ident: ref561 article-title: History-adjusted marginal structural models for estimating time-varying effect modification publication-title: Am J Epidemiol doi: 10.1093/aje/kwm232 – volume: 10 start-page: 24 year: 2005 ident: ref511 article-title: An experimental design for the development of adaptive treatment strategies publication-title: Stat Med – volume: 65 start-page: 331 year: 2003 ident: ref171 article-title: Optimal dynamic treatment regimes publication-title: J R Stat Soc Ser B doi: 10.1111/1467-9868.00389 – volume: volume 179, start-page: 189 year: 2004 ident: ref201 article-title: Optimal structural nested models for optimal sequential decisions publication-title: Proc Second Seattle Symp Biostat doi: 10.1007/978-1-4419-9076-1_11 – volume-title: Targeted learning: causal inference for observational and experimental data. year: 2011 ident: ref721 – ident: ref11 article-title: Information recovery and bias adjustment in proportional hazards regression analysis of randomized trials using surrogate markers publication-title: Proc Biopharm Sect – start-page: 69 volume-title: Latent variable modeling and applications to causality year: 1997: ident: ref431 – volume: 66 start-page: 688 year: 1974 ident: ref61 article-title: Estimating causal effects of treatments in randomized and nonrandomized studies publication-title: J Educ Psychol doi: 10.1037/h0037350 – start-page: 39 year: 2013 ident: ref381 article-title: van der Targeted data adaptive estimation of the causal dose response curve Technical Report Division of Biostatistics University of California Berkeley submitted to JCI |
| SSID | ssj0041819 |
| Score | 2.3758042 |
| Snippet | We consider the estimation of an optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment,... |
| SourceID | pubmedcentral proquest pubmed crossref walterdegruyter |
| SourceType | Open Access Repository Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 305 |
| SubjectTerms | Biostatistics - methods causal inference cross-validation dynamic treatment loss function Models, Theoretical oracle inequality |
| Title | Super-Learning of an Optimal Dynamic Treatment Rule |
| URI | https://www.degruyter.com/doi/10.1515/ijb-2015-0052 https://www.ncbi.nlm.nih.gov/pubmed/27227726 https://www.proquest.com/docview/1860825829 https://www.proquest.com/docview/1792378960 https://pubmed.ncbi.nlm.nih.gov/PMC6056197 |
| Volume | 12 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1Lb9QwEB5BKyQuFW8CpQoS4oTVxM_khHi0qpAoqLTS3iI7sduikl22G6H-e2YSJ7AUOHuixDO25xvP5BuAF0IEb3PnmVPUwkwJw6wsNVOulioPhbIl_Y388VAfnMgPMzWLF26XsaxyPBP7g7qZ13RHvpsXmqKZgpevF98ZdY2i7GpsoXETNom6jEq6zGwKuCR6L4K_uCklU0bMIscmuvDd868OF0iuGN2Lrvuka0Dzer3k1o8-l93402V3tRpzp71L2r8DWxFLpm8G49-FG769B7eG7pJX90F86RZ-ySKD6mk6D6lt0094RnzDp94PrejT47HSPD3qLvwDONnfO353wGKPBFYjUFsxI71qRMZD3pgmc8YXweReOSN0CEbUJUKQWtbSyVB7zr3RtsiVbRxvrEYp8RA22nnrH0NqPEKlLLNFyXOZG4SOGl2V4yJkNtOFS-DVqKWqjgTi1MfioqJAApVaoVIrUmpFSk3g5SS-GJgz_iW4Paq8ihvosvpl7gSeT8O49CmfYVs_71CGuA9NgTFYAo8GC01v4oZzDBx0AmbNdpMA0Wqvj7TnZz29tqagqjQJyD-s_NvX_W0eRI6XqSf_n81TuI0P6KFkchs2VsvOP0NYs3I7_drdgc23e4efj34C7Vf2TA |
| linkProvider | ProQuest |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3db9MwED-NTgheJr7JGBAk4Alr8XfygBCwTR3bChqd1LfMTpzRaUtL12jqP8XfyLlJCmXA255zTuK7893vfPYdwEvOC2eodcRK38JMck2MSBSRNhOSFrE0ib-NfNBT3SPxaSAHK_CjvQvjj1W2NnFuqPNR5vfIN2msfDQTs-Td-DvxXaN8drVtoVGrxZ6bXWLIdvF2dwvl-4qxne3-xy5pugqQDKHNlGjhZM4jVtBc55HVLi40ddJqropC8yxBp52JTFhRZI4xp5WJqTS5ZblRSMXxvTdgVXAMZTqw-mG79-Wwtf0C_aUH3GgGBJGaD5qqnggaNoenFlWSSuJ3Ype94BVoe_WE5trlPHueu5NJNZu22dq5E9y5A2sNeg3f1-p2F1ZceQ9u1v0sZ_eBf63GbkKamq0n4agITRl-Rqt0jqO2ZqU5H2Zhvz3bHh5WZ-4BHF0L_x5CpxyV7jGE2iE4iyITJ4wKqhGsKnSOlvEiMpGKbQBvWi6lWVOy3HfOOEt96IJMTZGpqWdq6pkawOsF-biu1fEvwo2W5WmzZC_SXwoWwIvFY1xsPoNiSjeqkMZXW9QxRn0BPKoltPgS04xhqKIC0EuyWxD4Qt7LT8rht3lBb-XDuEQHIP6Q8m9_97d5-HJ8kVz__2yew61u_2A_3d_t7T2B2zhY1Qc2N6AznVTuKYKqqX3WaHIIx9e9eH4Ccykz5g |
| linkToPdf | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1Lb9QwEB7BVqBeyrsNFAgS4oS7iZ_JsdAu5VUQtGhvkZ3YbaFkV9tEVfn1jJuHdrtwgbPHiT2e8czY428AnjPmrI6NJUb4EmaCKaJ5KokwORexS4RO_Wvkj_ty75C_G4vx3Ct-n1ZZ2KNZfVE1CKnDYpLX_qCsxxpACzw8-W5wfWNB_LHmcFq467CScAxlBrCy_ebV7rduN-ZowbwLjIrJiVBs3OJsLn1k0S4tOZvLOZNr55f32f1g58zS6BbobkJNNsqPrboyW_mvK1iP_zPj27DW-qzhdiNkd-CaLe_CjaaK5cU9YF_rqZ2RFqn1KJy4UJfhJ9yLfmKvnabkfXjQZbSHX-pTex8OR7sHr_dIW4uB5OgQVkRxKwoWURcXqoiMsolTsRVGMemcYnmKrk7Oc264yy2lVkmdxEIXhhZaIhV7AINyUtoNCJVFlyyKdJLSmMcKXVSJJtFQ5iIdycQE8LJbiSxvgcp9vYzTzAcsyIsMeZF5XmSeFwG86MmnDULH3wg3u2XNWkU9y-JE-iA5oWkAz_pmVDF_b6JLO6mRxmMsqgRjvQDWGyno_0QVpRigyADUgnz0BB6-e7GlPDm-hPGWPnhLVQD8iiTNje5P8_AgfJF4-G_dnsLNzzuj7MPb_fePYBUbZZO9uQmDalbbx-hhVeZJq0S_Af3EIEk |
| 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=Super-Learning+of+an+Optimal+Dynamic+Treatment+Rule&rft.jtitle=The+international+journal+of+biostatistics&rft.au=Luedtke%2C+Alexander+R.&rft.au=van+der+Laan%2C+Mark+J.&rft.date=2016-05-01&rft.issn=2194-573X&rft.eissn=1557-4679&rft.volume=12&rft.issue=1&rft.spage=305&rft.epage=332&rft_id=info:doi/10.1515%2Fijb-2015-0052&rft.externalDBID=n%2Fa&rft.externalDocID=10_1515_ijb_2015_0052 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2194-573X&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2194-573X&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2194-573X&client=summon |