dynamic dividend model - florida international …：动态的股利模型-佛罗里达国际…

《dynamic dividend model - florida international …：动态的股利模型-佛罗里达国际…》由会员上传分享，免费在线阅读，更多相关内容在教育资源-天天文库。

1、EmpiricalMethodsinFinanceandAccountingBong-SooLeeUniversityofHoustonAugust2004TABLEOFCONTENTSI.IntroductionII.DynamicdividendbehaviorIII.PermanentandtemporarycomponentsIV.Post-earningsannouncementdriftV.MeanreversionVI.CausalityVII.VARframeworkVIII.PresentvaluemodelsIX.GMMX.Thetermstructureofi

2、nterestratesXI.AlternativetestsofthemarketefficiencyXII.TimeSeriesModelsofHeteroskedasticity186ApplicationsandExamples[Motivation]1.Dynamicdividendmodel·Thepartialadjustmenthypothesis,whereDt*=targetdividend,EarningsPermanentcomponentofearnings:presentdiscountedvalueoffutureearnings2.Meanrever

3、sionbyFama/French(1988)Whatwillbetimeseriesmodelofpt?3.Market’sresponsetounexpectedearningsHowtoincreaseRsquareandλ?4.Post-announcementdriftinabnormalreturns186HowtounderstandserialcorrelationinARt?2.Rationalexpectationseconometrics·Economics/Finance/AccountingTheoryrestrictionsondata·Causalit

4、y[e.g.,earningsandinvestment]·VAR(Vectorautoregression)andidentifications[e.g.,fadorbubble]3.MarketefficiencyGMM186I.INTRODUCTION1.NOTATION·Astochasticprocessisacollectionofrandomvariable,acollectionindexedbyavariablet.·Lagoperator·PolynomialintheLagOperatorExample:186·Astochasticprocess{yt}is

5、stationary,iffforanysetofAofpaths,Pr.(A)=Pr.(LsA)foranyS.®Thesameprobabilitydistributionprevailsinanytimeperiod.[orifthejointdistributionassociatedwithkobservatonsmadeatanysetoftimet1…tkisthesameasthejointdistributionofforallhÎT](NOTE)t1…tkarenotnecessarilyconsecutive.·Astochasticprocess,yt,i

6、scovariancestationary,if(Note)Cy(t)=covariogramdependsonlyont,notont(ors)Cy(0)=var(yt);Cy(t)=Cy(-t):symmetricEX:Cy(1)=cov(yt,yt-1)=cov(yt-1,yt)=Cy(-1):anapplicationoftheSchwarzInequality186·etiswhitenoise(W.N.)[¹i.i.d.]if-i.i.d.isastrongerassumptionthanW.N.-etiscovariancestationary:constantmea

7、n-covariancedependsonlyont1862.COVARIANCEGENERATINGFUNCTION:gy(z)-MomentGeneratingFunctiontellsuswhethertwovariablesarefromthesamedistribution.·ConsiderarandomvariableytwhereetisW.N.andisassumed(toassurevar(yt)isfinite)(Note)·Covariance