the cross-section of stock returns and cash flow risk

《the cross-section of stock returns and cash flow risk》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。

1、DoctoralSeminarinEmpiricalFinanceTopic6TheCross-SectionofStockReturnsandCashFlowRiskLecturer:LarsA.LochstoerLondonBusinessSchoolJanuary31,20081Wheredobetascomefrom?AssumealinearSDF(canbeobtainedbylinearizinganymodel)withKfactors,Ft–Thismodelimpliese0EtRit+1=itt:–The’saretheassetreturn’scon

2、ditionalsensitivitytotheriskfactors,Ft.Inthistopic:thinkaboutthedriversofrealizedassetreturnsandwheretheco-movementfromthefactorscomefrom–Commonmovementsindiscountrates?–Commonmovementsincash‡ows?Letmeknowofanyerrors,please.ThesenotesdrawontheexcellentPhDcourseinempirical…nanceItookatBerkeley,

3、taughtbyGregoryR.Du¤ee.Contactinfo:LarsLøchstøer,P222,LondonBusinessSchool,SussexPlace,Regent’sPark,NW14SA,London,UnitedKingdom.E-mail:LLochstoer@london.edu0Campbell’s(1991)decompositionforassetreturnsCampbell-Shillerdecomposition:P1P1j1j1ptdt=k+
dt+jht+j:j=1j=1Solveforlogreturnint+1P1P1

4、j1j1ht+1=k(ptdt)+
dt+jht+j:j=1j=2Takeexpectationsasoftandt+1P1P1j1j1Et[ht+1]=k(ptdt)+Et[
dt+j]Et[ht+j]:j=1j=2P1P1j1j1ht+1=k(ptdt)+Et+1[
dt+j]Et+1[ht+j]:j=1j=2Subtracttheformerfromthelatter:P1P1j1j1ht+1Et[ht+1]=(Et+1Et)
dt+j(Et+1Et)ht+j:j=1j=2Shorthandlanguage:–New

5、saboutcurrentandexpectedfuturecash‡owsP1j1d;t+1=(Et+1Et)
dt+jj=1–NewsaboutexpectedfuturereturnsP1j1h;t+1=(Et+1Et)t+1ht+jj=2Innovationinassetreturn(withtilde)isdi¤erencebetweennewsaboutcash‡owsandnewsaboutreturnsh~t+1=d;t+1h;t+1Wecaninterpretthisequationineitherrealornominalterms.1W

6、ecanfurtherdecomposereturnsintorisk-freereturnandexcessreturn:ht+1=rt+1+et+1Impliescorollarydecomposition(noteindexes)P1j1et+1Et[et+1]=(Et+1Et)t+1
dt+jj=1P1P1j1j1(Et+1Et)rt+j(Et+1Et)et+j:j=2j=2I.e.,innovationinexcessreturnisnewsaboutcash‡ows,lessnewsaboutfuturerisklessratesandfutur

7、eexcessreturnse~t+1=d;t+1r;t+1e;t+1:ImplementingtheCampbell(1991)decompositionusingaVARAssumethatlogreturns,logD/P,perhapsothervariablesfollowaVAR:01pt+1dt+1BCzt+1=@ht+1Axt+1zt+1=Azt+~zt+1;z~t+1N(0;)Estimatemodelin