HeapinganditsConsequencesforDurationAnalysis Sonderforschungsbereich386Paper2032000 Onlineunterhttpepububunimuenchende Projektpartner Thispaperanalysestheconsequencesofheapingindurationmod ID: 101277
Download Pdf The PPT/PDF document "Wolff,Augustin:" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Wolff,Augustin: HeapinganditsConsequencesforDurationAnalysis Sonderforschungsbereich386,Paper203(2000) Onlineunter:http://epub.ub.uni-muenchen.de/ Projektpartner ThispaperanalysestheconsequencesofheapingindurationmodisaspecicformresponseerrortretrospeclaborRespondenspellepisodebasedCalendarbasedquestionnairesinsteadmayleadtoabnormalconcentrationsofthestartandorendofspellsatspeciccalendarypeingwhicKrausandSteiner tiedforspelldatafromtheGermanSocioEconomicspecialcaseofanexponentialmodelheapingwithasymmetriczerodoesmodeldependencediscusstheproposalberstudieswkthetheoreticalresultsTheMonteCarlosimalsoshowthatanamountofheapingthatcharacterizestheGSOEPdoeseibullmodelitclearlyleadstospuriousseasonalsomedirectionsoffutureworkareindicated SeminarfurempirischeWirtscungLudwigMaximiliansUnivyofMuhLudwigstrRGD MunichJoacademiestrIDMunichthomas statunim oltroductionRetrospectivrespondentoprovideevthistoriesoflaborforcestatesordurationofsomelaborforcestatesubstantialresponseerrorsmayemergeHeapingorroundingoisaparticularformofsucherrorsTherearetypesofheapingdependingonthedesignofthequestionnaireemergefromcalendarbasedSocioEconomiccalendariumrequiresrespondenpossiblelaborHeapingimpliesthatrespondenroundooruserulesofthreportinglaborrespondenwhobecomeunemploinFebruaryorMarcjustreportthestartoftheirspellheapingpatternforunemplotdurationdataoftheGSOEPWresponsesepisodebasedquestionnairesWhenrespondentsrecalltheentiredurationofspelltspellsfromItalianLaborForceSurveydataaresubjecttothisheapingBoththestudyofTorelliandTellatoaswellasthatofKrausandSteinerproposedmodelstestimatesoftheparametersinthepresenceofheapingKrausandSteinerwhodealtwithheapingincalendarbasedquestionnairesprovidednoinformationaboutthegeneralconsequencesfordurationanalysisofthetypepossibleexponenmodelloglogisticmodelspecic ThereissomeevidencethattheeastGermanunemplotspelldataoftheGSOEParealsocharacterizedbyheapingWolercomparedwiththendingsofKrausandSteinerforthewestGermanunemplotspelldatathisamountofheapingislo olprocessItisawwnresultthatazeromeanmeasurementerrorofthedepenindependendoesmodelsresponsespellsyadvestimationresultsdurationmodelsmodelsheapedaimofthispaperistoulationstudiesreticalconsiderationswhetherandwhenspecicformsofheapinghaThisisparticularlyimportanttoknoifthereisnosucienpaperheapinginlaborforcesurvSectionthreeshowswhetherheapingevenasspellparameterestimatesifthetruespelllengthsaredrawnfromtheexponenulationstohighlighttheconsequencesofheapingforparameterestimatesSectionvesummarizesourverypreliminaryresultsandindicatesextensionsofthisworelliandellato studiedLaborinformationof TheItalianLaborForceSurveyisaquarterlysurveyofrotatingpaneldesignfamilyisinedfortoconsecutivesurveysandthendroppedfortosurveysandedagainfortonaleoplewhoconsiderthemselvesasunemploersareaskedforhowmanymonthstheyhaealreadybeenlookingforajob olresponsesretrospectivlengthofanunemplotspellinprogressatthedateofthesecondsurvrstandsecondof Itconsistsofindividualsagedbeteen and yearswhoareunemploedintherstsurvThissamplevidedevidenceforabnormalconcentrationsofunemplotdurationatpercenreportedspellultiplesof monthsformenandwomenrespectivItisthusclearthatrespondentsusedaruleofthbtoreportthelengthoftheirunemplospellstheirdurationofunemplotomultiplesofsixor durationtotheseabotdurationTheauthorspointedoutthattherearetruebehaioralreasonsthatcouldleadtosuchspikesatsimilarspelllengthsTheymafromthetimeunbenetsforstudiesofunemploTheauthorsdevelopedamodelthatyieldsconsistentestimatesofacontimeparametricdurationmodelinthepresenceofheapingLetTbeaconuousrandomvariableofdurationofunemplotwithprobabilitydentthatcompletedspelldurationsareobservedfornsubjectspelllengthbheapedMarrangedinincreasingorderisaBernoullirandomvariablewhichequalsoneforheapedvaluesandzerootherwiseLetfurther betheythat where isaparametricfunctionknouptodescribesthetruespelllengthTheinferentialproblemistoestimatefromthedatathataresubjecttoheapingandaknosetofheaped olelihoodofarandomsampleiszznonheapedheapedupperheapedspelllength beingconstantheinalselihoodbecomes isdistributionfunctionfactorisingthelatterlikelihoodcomponenmodelolvingtheheapingfunctionemergetheparameterscouldbeconsistentlyestimateddisregardingtheheapingproInordertorevealtheeectsofheapingtheauthorsgenerateddurationdatathatstemfromtheexponentialtheWeibullandtheloglogisticdistributionTheydidnotintroducecoTheresultingspelllengthswerealteredbaheapingpatternthatisclosetotheonethattheyrevealedfromtheItalianLaborexponenelihoodfunctionselihoodofmodelsexponenmodeldoesloglogisticdurationmodelsereconsiderablyapartfromtheirtruevalueswherethebiasappearstobepositivelyrelatedtotheamountofheapingNexttheyconcludedthatacrudeelihoodeparameterestimatesconsiderablyInthecaseofalargesamplesizemodelselihoodestimatesclosetotheirtrue olspellsrespectretrospectivlaborrespondencodehisherlaborforceoccurpeopletooatcertaincalendarmonthsandnotfrequentlyenoughatthemonthsthatareclosetothemThestudyofKrausandSteineridentiedsuchabnormalconanalyzedaninowsampletoregisteredunemplotheGSOEPWestfromJanuary toDecember eoplewhowsampleofuncensoredandrightcensoredunemplotspellsThesepeoplebeenbehafromunemplowsaLaborbothationperiodThestrikingndingsweretoabnormalconcenpopulationperiodspondingpopulationvComparedwiththeirpopulationvaluestheoutowratesoftheGSOEPberboringmonthsOctoberandNobertendtobesomewhatloThesetoheapingpatternsimplythatspelllengthsarereportedastoolongforspellsthatstartintherstquarteroftheyearandorterminateduringitslastquartereconometricdurationmodelsnegativdependencemaybeoerstatedandbiasedcoecientsforseasonaleectsma TheauthorsaccountedforsampleattritionoftheGSOEPWestrelativetotheregisterdatabyappropriatewtingfactorscalculatedonayearlybases olulationstooftheidentiedheapingpatternforstandarddurationmodelsNordidtheydealwiththistopicfromatheoreticalperspectivTheyadjustedthemodelofTorelliandTellatoinordertoincorporatethespecicheapingpatternthattheyspellsofthetoadiscretetimeproportionalhazardsmodelspellsmodelspeciedberspelltheheapingprobabilitiesusingoutsideinformationKrausandSteinerproceededthenbycomparingtheestimatesofseveraldurationmodelsthattakheapingintoaccountorcompletelyignoreitAllmodelsincludedastandardellasabaselinehazardLetussummarizetheirresultsTheauthorsfoundhardlyanydierencebeproportionalhazardmodelwithexiblebaselinehazardwithandwithouttheircorrectionforheapingincorporatedelihoodcoefmodellargethesameasforthoseofthehazardmodelwithaexiblebaselinehazsmootherberproportionalmodelincludingdummiesbercoecienmodelsdummiesandynotbeduetoheaping TheauthorsagainusedinformationaboutthemonthlypopulationinowandoutoaspublishedbytheGermanFederalLaborOceTheseincludedageforeignerdisabilitmaritalstatuseducationhouseholdincomeandtheregionalunemplotrateTheychosealogittransformationoftimenamelyexpexpastheirbaselinehazardfunction ollookmodelInthissectionwetakeasimpliedlookatnoncorrectedmaximumlikelihoodmodeldistributedwithparametersifitsdensitexpdependsdependencyypothesis correspondstomonotonelyincreasinghazardcontainingtheRadistributionwithlinearhazard Thespecialcaseofconstanthazard istheexponentialmodeltroducesoneusuallyparameterizesexpThentheWeibullmodelisaspecialacceleratedfailuretimemodelKalbeiscPren p amodeloftheindependentlyandidenticallydistributedHeretoobtaintheWmodelistakasextremevdistributedandissetAppendixspecialcaseofahomogeneoussamplewithoutindividualcoariatesienerthelessusethereparameterizedform insteadofitselfbecauseitis oltheusualwyoflookingattheWeibullmodelineconomicsitallowstouseforageneralizationofourtocoEstimationoftheunknownparametersistypicallydonerelyingonumlikelihoodprinciplecorrespondingexp expprocedureisneedederifonetreatsasxedtherstlineyieldsanexplicitsolution Pni Ti elihoodelihoodheapedadditionalconsiderationsareneededheapedelihoodoformallyintroduceaheapingmechanismtypicalforcalendarbasedquestionnairesweassumethateveryspellmaybeheapedwithacertainprobabilywhichisassumedtobeindependentofthecoariatesandthespelllengthDenoteforsomesucientlylargetheprobabilitythataspellisprolongedbqandbtheprobabilitythatthespellofthedurationtimesTT olTTdependonthemonthofenAsshowninAppendixAassumingindepenbehabelowdoesonlydependonthemarginaldistributionoftheThereforeaftercalculatingthemarginalheapingprobabilitiesproceedlossofgeneralitywiththemodeldescribedin plugginginheapedTT lnPni Ti goodqformallymabecomenegativDependingofthequestionnairethismaoftenbeunrealisticonesimplythenwouldnotrecordedthesespellsInthiscasetheheapedornaive maximumlikelihoodestimatew lnPniT max etacitlyassumethatiswelldenedThisisalwysthecasefororifequalszeroforev olberofspellswithumlikelihoodrespectthisarstexplorationofthebehaviourofsamplesizewingtoinnitywillbeperformedoobtainarstimpressionoftheasymptoticpropertiesofandofrestrictourattentiontotoimportantspecialcasesnamelytheexponendistribution andtheRayleighdistribution TheasymptoticbiascanbegivaclosedProposition welldenefollowing expln exp ByusingtheGeneralBinomialtheoremtheconsiderationsgivenbelowcanbeeasilybeoddproceduretroubleifforsomeThisisthecaseforanyrealisticconstellationBytheirdependenceontheunknotheseadditionalconditionsmaybesometimesButnotethatitisalwyssatisedinthecaseofsymmetricheapingaswellasforthepositivelybiasedonesidedheapingpatternobservedintheGSOEP ol exp qXl ll l exp BBBXl ll l exp qXl ll l expAppendixThebiasgrowsatterbythelogarithmicfunctionNotefurtherthatthebiasproportionalspelllengthsmallerceterisbecomesspellwillshoproposalestimationofBeforediscussingthiswebrieywttolookattheestimateasdenedin andthebehaofbothestimatesinsomespecialcasesProposition proimmediatelyaboundbiasoftheemainvalid ifonerandequalitiesbytherelationsgr olReturningtooextremeybeofspecialinTherstoneistheconstellationwhereintheypeforallulaeaboThesecondoneisthesymmetricsituationwhereforevtheproportionofthespellsprolongedbandtheproportionofthespellsshortenedbypicallywillnotIfinthesituationofProsition theheapingis i e forallq thenisconsistentinthecaseoftheInthesituationofCorollaryoneregularlyhasexponensmallbiasexpectedputtingnegativtozeroTheTorelliandTellato caseisactuallynearlysymmetricSoCoroltionsdiscussedinSectiontInanexponentialmodelthebiasfromignoreibullmodelwithdurationdependenceqestimationpropossible olProposition expln exp qXl l l lvuut Appendixtheresultsgainedbytheseconsiderationsareofcoursequitepresharpeningboundnaturenaturallyprovidingnoconcretestatementonnitesamplebiasdependencetedherenitesamplesizeastronglationbetestimatesofwillbewingsimulationstudy oldoesmodelserrorisofthezeromeansymmetrictypecouldasymptoticallyleadtobiasedestimatesoftheparametersoftheWeibullmodelInthissectionwepresenresultsfromsimulationsinordertostudythenitesamplebehaviourandtheallwypemodelssimilartospellaboutexpectpopulationSecondweattempttoshowwhethertheparameterestimatesarelessprecisetroducesdependenceeusetheestimatingequations and derivedfromtheideallikelihoodheapedulationrunstheestimationprocedure timeseassumethatspelllengthsmeasuredinconuoustimetheunitismonths andthattherenocensoringypetionstudiesconsiderdiereneibulldistributionsasDGPsofthespellspellspecicspellheapedpercenspellbercorrespondingspellpercenspellberbecome oldurationispercenheapedspellsaboutpopulationspellsulationsforbetAlsothedurationdependenceisvdurationdependenceparameterthatresultfromestimationwiththeheapedThecorrespondingaeragestandarderrorsandthosethatareacyMLestimationwiththeoriginaldatapriortoheaping arealsodisplaanela ulationresultswithoutandurationdependenceexponenspellbothdependenceheapedbothcoecienthedurationdependenceparameterthatresultfromtheheapeddatasoforahigheraeragespelllengthThemeancoecientsinthepresenceheapingarevtotheparametervInPanelb theunderlyingDGPisaWeibulldistributionwithpositivedurationdependenceThemeanparameterestimatesinthistablehardlydierfromtheparametersoftheDGPNeitherarethesimulationresultsforheapedoriginaldataecarriedoutthesameanalysisforaWeibullDGPwhereissettoieedurationdependenceTheresultsaredisplaedinpanelc ofTable HereweclearlyseethatthemeanestimateoftheconstantfortheheapeddataexceedssomewhatitstruevAlsotheaerageduration ThistypeofheapingleadstosomenonpositivespelllengthsEgaspellthatlastsforonemonthofwhichthestartisheapedfromDecembertoJanuarywouldneverhabeenobservSowediscardedsuchspellsintheheapedsampleThismeansweanalyzethebehaviouroffromandnotthatoffrom olspellpercentofthespellstartsheapedfromDecemberandFebruarytoJanpercenspellsstartsheapedfromNoberanddependenceexponenspellberofobsheapeddataoriginaldata heapeddataoriginaldataEstimatedmeanEstimatedmeanCoe Coe true positivdependence berofobsEstimatedmean Estimatedmeanheapeddataoriginaldata heapeddataoriginaldataCoe Coe true true dependence berofobs heapeddataoriginaldata heapeddataoriginaldataEstimatedmeanEstimatedmeanCoe Coe true berofsimulations oldependencecoecientishigherthanthetrueNoneofthesedierencestaccordingtotheaeragestandarderrorsNextthereasonforthesymmetricheapingspellsoriginallengthpletelydiscardedduetotheheapingTheproportionofsuchspellsincreasesspellberspellstheislikelytobecomesomewhathigherthanitstruefarissymmetricheapingspellsheapedmodelspellAppendixreectscharacteristicsofarealworldsampleofunemplotspellspoinparameterestimatesspellwingsimSupposespellexponentiallydistributedsothereisnodurationdependenceSomeresponspellsbetspellreportedspellspellexpectedeibulldistributionwiththeheapeddatatheestimatedconstantshouldbeupardbiasedpositivdependencetimewsamplesizeandofheaping Theseresultsareaailableonrequest olspellsstartinJanuarybyonemonthwhileofspellsthatstartinFebruaryareincreasedbomonussomewhatlessthanoftheoriginalspellsareincreasedinlengthAgainwecarriedout simulationsestimatingthebothdependenceexpectedsamplesizen InPanelb ofTable weraisedthemeasurementerrorsubstanpercenspellspercenpercenheapedtothestartingmonthJanmorethan ofthespellshadtheirspelllengthscTheresultisthatboththeadependencetheheapeddataincreasedinsizeascomparedtoTable a Wherethetruetisequaltoonetheaerageestimatedconstantis totheestimatedmeanstandarderroritissignicantlydierentfromitstruepositivdependencebecomesmodelspellestimatedparametersremainlargelyunalteredfromthehooseexponenpatternwillbesuchthatonaeragepercentofthespellsthatendinOctoberareprolongedbomonthsandpercentofthosespellsthatendinberwillbecomeonemonthlongerSotheirspellendissettoDecemberexponenofco olableulationresultsforexponentiallydistributedspellspercenspellheapedpercenspellheapedheapeddataoriginaldata heapeddataoriginaldataEstimatedmeanEstimatedmeanCoe Coe true percenspellheapedpercenspellsheapedpercenspellsheapedheapeddataoriginaldata heapeddataoriginaldataEstimatedmeanEstimatedmeanCoe Coe berofobsberofsim olexpospellpercenspellheapedberberpercenspellheapedOctoberberHeapedEstimatedmeanCoe oct berofobservberofsimaryingdummariablesforfromJanOctoberberberableshowsoursimulationresultswhenestimatingtheparameterswiththeheapeddataTheyclearlysuggestthatthereisaspuriousseasonalcoecienOctoberberimpliesthatbeloberbercoecienandsignicaninthismonthistoohightiedeectsofheapingontheparametersestimatesofsomemodelsthatisusualintheGSOEPWestisconsideredwereachedthefollowingcon oldoesdarderrorsthatdierfromthoseestimatedwiththeoriginaldataheapingpatternsofthekindthatKrausandSteinerfoundfortheunemplotdurationdataoftheGSOEPWestarenotsymmetricTheirvstudyfoundthatrespondentstendtoplacethestartoftheirspellstoooftentoJanuaryandnotfrequentlyenoughtoFebruaryandMarcSothespellsmodelypepositivdependencebecomespellypothesesspellsheapedspecicoccurexponenaneectiftheisonlyofaboutofspellsThereareseveralpossibleextensionsofthisworkwhicttoaddresstheoreticallyaswellasbysimFirstitisplausiblethatheapingmadependSupposerespondenbecomeedregularlyataroundthesametimeoftheyearseasonallyunemploOnemayexpectthattheyaremorelikelythanotherstoheapthespellsbecomeunemplofocusailableinmanycounThereforetheuseofoutsideinformationshouldpossibleincorporateindependenelihoodmethods OneotherwytocorrectforheapingcouldbebyincludingasetofdummforthestartingmonthsascoResultsthatareaailableonrequestsuggestthatthisindeedimproestheestimatesoftheconstaneritleadstoalargerpositivbiasofthedurationdependenceparameter olmethodstofheapingasypeofinalcensoringearegratefultoHansforhelpfuldiscussionsandJDandRL TheStatisticalANewModellingtdurationmodelswithanapplicationtoretrospectivSocioEconomicPModellinginaccuraciesdurationdataEuropeanAppendixProofPropositionproofPropositionberandonthefollowinglemmaWeibullexp exp oloccurring du dt toGammain exp u troducestochasticallyindependentofdescribingtheheapingsuchthatAccordingto onehasfor nIE TiE TiHiE TiE Hiexp Xl l llThereforeplimn plimnBBBBBBBBnXl Ti nCCCCA CCCCAlnBBBBexp BBBBXl l l l CCCCACCCCA lnBBBBXll l l exp CCCCA ol qXl l llqXll llexp BBBBXl l l l exp p qXll l l exp CCCCAThereforelimn plimnBBBB lnBBBBnXi Ti nCCCCA CCCCA BBBBXl l l l exp p qXll l l exp CCCCA BBBXl l l l exp p qXll l l CCCA olAppendixProofPropositionromProposition Parta exp exp exp exp ThereforebySlutzkysTheorem olromProposition Partb BBBBXll l l exp p qXl l l l exp plim Xl l l l exp qXll l l exp exp exp plimUsing theassumption exp madeinPropositionimpliesplimThisguaranquadraticformintheabopossessesawelldenedanduniquesolutionnamelyexp p qXl l l lvuu plim ol qXl p l l lvuut qXl l l l qXl l l l A qXl l l lvuut plimAppendixDependsdependforeverymonstheheapingprobabilitiesljljdescribingtheentrymonthofunitAssumethatthetruedurationisstochasticallyindependentofandanalogoustoaboethatareconditionallyindependentgivurthertheenistakentobeindependentlyandidenticallydistributedamongallunitsninthesamewItisshownthattheexpectationdependsonlyonthemarginalprobabilitiesljqljq olAppendiypendixAtodiscussbiasandbiascorrectionarealsovalidforthecaseconsideredljljljljljljljljAppendixAppendixberisintroducedtoanexponentialDGPTheheapingisagainsuchthatthereispercenspellberapercenthatthespellstartsberwillbecomeJanTheprobabilitythataspellstartsinaspeciccalendarthisnolonger Thecalendarstartofthespellaredrawnsuchthat olableBulationresultsforexponentiallydistributedspellsincludpercenspellheapedberpercenspellsheapedberEstimatedmeanCoeagehnicalunivydegreetRatioberofobsberofsimwiththedistributionofthepopulationinoercalendarmonthsintoregisteredunemplotinEastGermanertheperiod to coecienandstandarderrorsof simulationsofmaximumlikelihoodestimationofelihoodulatedparametertlydierenfromtheirtrue Wolff,Augustin: HeapinganditsConsequencesforDurationAnalysis Sonderforschungsbereich386,Paper203(2000) Onlineunter:http://epub.ub.uni-muenchen.de/ Projektpartner