资源描述：
《Large Sample Properties of Generalized Method of Moments Estimators》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。
1、LargeSamplePropertiesofGeneralizedMethodofMomentsEstimatorsAuthor(s):LarsPeterHansenReviewedwork(s):Source:Econometrica,Vol.50,No.4(Jul.,1982),pp.1029-1054Publishedby:TheEconometricSocietyStableURL:http://www.jstor.org/stable/1912775.Accessed:14/03/201221:45YouruseoftheJSTORarchiveindica
2、tesyouracceptanceoftheTerms&ConditionsofUse,availableat.http://www.jstor.org/page/info/about/policies/terms.jspJSTORisanot-for-profitservicethathelpsscholars,researchers,andstudentsdiscover,use,andbuilduponawiderangeofcontentinatrusteddigitalarchive.Weuseinformationtechnologyandtoolstoin
3、creaseproductivityandfacilitatenewformsofscholarship.FormoreinformationaboutJSTOR,pleasecontactsupport@jstor.org.TheEconometricSocietyiscollaboratingwithJSTORtodigitize,preserveandextendaccesstoEconometrica.http://www.jstor.orgEconometrica,Vol.50,No.4(July,1982)LARGESAMPLEPROPERTIESOFGEN
4、ERALIZEDMETHODOFMOMENTSESTIMATORS1BYLARSPETERHANSENThispaperstudiesestimatorsthatmakesampleanaloguesofpopulationorthogonalityconditionsclosetozero.Strongconsistencyandasymptoticnormalityofsuchestimatorsisestablishedundertheassumptionthattheobservablevariablesarestationaryandergodic.Since
5、manylinearandnonlineareconometricestimatorsresidewithintheclassofestima-torsstudiedinthispaper,aconvenientsummaryofthelargesamplepropertiesoftheseestimators,includingsomewhoselargesamplepropertieshavenotheretoforebeendiscussed,isprovided.1.INTRODUCTIONINTHISPAPERwestudythelargesampleprop
6、ertiesofaclassofgeneralizedmethodofmoments(GMM)estimatorswhichsubsumesmanystandardecono-metricestimators.Tomotivatethisclass,consideraneconometricmodelwhoseparametervectorwewishtoestimate.Themodelimpliesafamilyoforthogonal-ityconditionsthatembedanyeconomictheoreticalrestrictionsthatwewis
7、htoimposeortest.Forexample,assumptionsthatcertainequationsdefineprojec-tionsorthatparticularvariablesarepredeterminedgiverisetoorthogonalityconditionsinwhichexpectedcrossproductsofunobservabledisturbancesandfunctionsofobservablevariablesareequatedtozero.Heuristically,iden
