资源描述：
《Large Scale inference_Bradley Efron_2010.pdf》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。
1、ThispageintentionallyleftblankLarge-ScaleInferenceWeliveinanewageforstatisticalinference,wheremodernscientifictechnologysuchasmicroarraysandfMRImachinesroutinelyproducethousandsandsometimesmillionsofparalleldatasets,eachwithitsownestimationortestingproblem.Doingthousandsofproblemsato
2、nceinvolvesmorethanrepeatedapplicationofclassicalmethods.TakinganempiricalBayesapproach,BradleyEfron,inventorofthebootstrap,showshowinformationaccruesacrossproblemsinawaythatcombinesBayesianandfrequentistideas.Estimation,testing,andpredictionblendinthisframework,producingopportuniti
3、esfornewmethodologiesofincreasedpower.Newdifficultiesalsoarise,easilyleadingtoflawedinferences.Thisbooktakesacarefullookatboththepromiseandpitfallsoflarge-scalestatisticalinference,withparticularattentiontofalsediscoveryrates,themostsuccessfulofthenewstatisticaltechniques.Emphasisisont
4、heinferentialideasunderlyingtechnicaldevelopments,illustratedusingalargenumberofrealexamples.bradleyefronisMaxH.SteinProfessorofStatisticsandBiostatisticsattheStanfordUniversitySchoolofHumanitiesandSciences,andtheDepartmentofHealthResearchandPolicyattheSchoolofMedicine.INSTITUTEOFMA
5、THEMATICALSTATISTICSMONOGRAPHSEditorialBoardD.R.Cox(UniversityofOxford)B.Hambly(UniversityofOxford)S.Holmes(StanfordUniversity)X.-L.Meng(HarvardUniversity)IMSMonographsareconciseresearchmonographsofhighqualityonanybranchofstatisticsorprobabilityofsufficientinteresttowarrantpublication
6、asbooks.Someconcernrelativelytraditionaltopicsinneedofup-to-dateassessment.Othersareonemergingthemes.Inallcasestheobjectiveistoprovideabalancedviewofthefield.Large-ScaleInferenceEmpiricalBayesMethodsforEstimation,Testing,andPredictionBRADLEYEFRONStanfordUniversitycambridgeuniversityp
7、ressCambridge,NewYork,Melbourne,Madrid,CapeTown,Singapore,SaoPaulo,Delhi,Dubai,Tokyo,MexicoCityCambridgeUniversityPressTheEdinburghBuilding,CambridgeCB28RU,UKPublishedintheUnitedStatesofAmericabyCambridgeUniversityPress,NewYorkwww.cambridge.orgInformationonthistitle:www.cambridge.or
8、g/9780521192491cB.
