资源描述:
《aer.100.2.16 Indicators for Dating Business Cycles Cross-History Selection and Comparisons》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、AmericanEconomicReview:Papers&Proceedings100(May2010):16–19http://www.aeaweb.org/articles.php?doi=10.1257/aer.100.2.16IndicatorsforDatingBusinessCycles:Cross-HistorySelectionandComparisonsByJamesH.StockandMarkW.Watson*TheNationalBureauofEconomicResearchusingmanyseries,rangingf
2、romtheearlywork(NBER)BusinessCycleDatingCommitteeandofRayFairandRobertShiller(1990)toworktheCenterforEconomicandPolicyResearchoverthepastdecadeonforecastingusinghun-(CEPR)BusinessCycleDatingCommitteedredsofseriesanddynamicfactormodels;seedatebusinesscycleturningpointsusingasma
3、llStockandWatson(2009)andSandraEickmeiernumberofaggregatemeasuresofrealeconomicandChristinaZiegler(2008)forrecentworkactivity.Forexample,initsmemorandumandStockandWatson(2010a)forasurvey.TheexplainingtheDecember2007peak(NBERproblemofdatingturningpointsdiffersfromBusinessCycleD
4、atingCommittee2008),thetheforecastingproblembecauseturningpointsNBERcommitteementionedthatitconsidersareestimatedretrospectively(in-sample)andfiveseries,quarterlyrealGDPandthe“bigbecausetheturningpointestimatorisnonlin-four”monthlyseries,realpersonalincomelessear,whereasthefor
5、ecastsconsideredinthetransfers,realmanufacturingandwholesale-many-seriesliteraturearepredominantlylin-retailtradesales,industrialproduction,andear.Ourapproachtodatingreferencecyclesisnonfarmemployment.(Theseseriesdonotinconceptuallyrelatedtotheapproachdevelopedgeneralreceiveeq
6、ualweight.)Incontrast,whenbyDonaldHardingandAdrianPagan(2006)theNBERresearchprogramondatingbusi-andstudiedbyMarcelleChauvetandJeremynesscyclescommenced,researchersexaminedPiger(2008);however,ourmethodsdifferinturningpointsinhundredsofseriesanddatedthedetails,andwefocusonusingm
7、anydis-businesscyclesbydetectingclustersofspe-aggregatedseries,somethingnotconsideredbycific-cycleturningpoints;seeArthurBurnsandthesetwopapers.Additionally,wetreatrefer-WesleyMitchell(1946,13and77–80).Thedat-encecycledatingasafrequentistestimationingofturningpointsevidentlyha
8、sshiftedfromproblemandprovidestandarderrorsforturn-aggregatin