基于现金流量的我国上市公司财务预警实证研究

云南财经大学硕士学位论文基于现金流量的我国上市公司财务预警实证研究姓名：杨瑶申请学位级别：硕士专业：会计学指导教师：李红琨2011-05I()()10(II)Logistic390A5Logistic50T-3T-2T-1ST69.7%74.9%78.9%STLogisticAbstractIIIAbstractAstheeconomicglobalizationdevelops,themarketchangesrapidlywhileuncertainfactorscomeupgreatly.Itcomeincreasingriskswithmoreandmoreopportunitiesforenterprises.Itmaycauseasignificantlossorevenbankruptcyforenterprisesifenterprisescannotpreventandcontrolriskseffectively.Withoutanydoubt,thefinancialriskisaveryimportantfactorofbankruptcyamongmany.Investorsandotherstakeholderscouldbesubjecttotremendouslosscausedbyfinancialrisk.What’sworse,anenterprisemaybeextinctbecauseoffinancialrisk.FinancialRiskForewarningaimstopredicttheoperatingandfinancialactivitiesanddiscoverthepotentialoperatingandfinancialriskduringthebusinessoperatingmanagementthroughfinancialstatementsandrelevantaccountinginformationprovidedbyenterprises.Itwarnstheenterprisemanagementleveltotakeeffectivemeasurestopreventpotentialrisksbecominglossesintimesothattotakeprecautionsbeforeitistoolate.Mostpreviousstudiesconductedmodelinganalysisbyusingthetraditionalmodeloffinancialindicators,rarelyrelatingcashflowindicators.However,theimportanceofcashflowanalysishasbecomecommonconcernincreasingly.Inthispaper,theauthoranalyzesthelimitationsoftraditionalfinancialindicatorsandtheadvantagesofcashflowindicators,whicharefoundtobemoreeffectivethanthetraditionalfinancialindicatorsonrespondingtotheactualfinancialsituation.Theintroductionofcashflowtocashflow-basedforewarningmodelsanalysisoffinancialpointofviewisnotonlyapplicableinpracticalactivities,butalsoinacademicstudy.Thispaperdiscussesthefinancialforewarningsystemtheory,theoryofcashflow,financialforewarningfunctionofcashflowandotherspecificthoughtstodeterminethescopeofindicatorstobeselectedwithin.Therefore,basedontheprincipleofindicatorsselection,thispaperestablishesafinancialriskforewarningsystembasedoncashflowindicators.Theauthorselects10indicatorsofcashflow,coveringthesolvencyofenterprises,capacityofearningcashflow,financialflexibility,developingcapacity.Throughadiscussandstudyofavarietyofexistingresearchmethods(single-variablemodel,multiplelinearmodel,multiplediscriminatemodel,multiplelogisticregressionmodel,AbstractIVmultivariateprobitregressionmodel,nonparametricanalysis,artificialneuralnetworkmodel,etc.),theauthorultimatelyselecttheprincipalcomponentanalysisandLogisticregressionanalysisintheempiricalandexperimentalstage.Intheempiricalandexperimentalstage,theauthorselects390A-sharelistedcompaniesofChinatostudy.Firstly,theauthormakespracticaluseofprincipalcomponentanalysistoselect5principalcomponentsthroughsievinganddimensionalityreducting.ThenLogisticregressionwasconstructedtobuildafinancialriskforewarningmodeloflistedcompaniesinChinabasedoncashflow.ByanalyzingandexaminingT-1,T-2,T-3-yearfinancialdataof50samplecompanieswhicharerandomlyselected,themodelisprovedtoberightasexpected.Moreimportantly,asSTtimeprogresses,themodel'saccuracyincreases.Theidentificationaccuracyratewas69.7%,74.9%,78.9%,whichmeanstheoccurrenceoffinancialriskisagradualprocess,thecloserSTgoes,thegreaterthefinancialrisk.Intheend,theauthorproposesprospectoffutureresearchdirections.Keywords:FinancialForewarning,CashFlow,PrincipalComponentAnalysis,LogisticRegressionAnalysis119901219STPT2()Fitzpatrick(1932)19\\Beaver(1966)19541964797930\\\\Beaver()Altman(1968)Zscore33335\\\\\3Z123451.21.43.30.61.0zxxxxx=++++AlimanHaldemanNarayanan(1977)ZscoreZETAEdmister(1972)N(0,1)()Martin(1977)LogitOhlson(1980)19701976105205896.12%()Zmijewski(1984)ProbitProbit4()1.Frydman,AltmanKao(1985)2.Laneetal.(1986)()CoastFant(1993)()(1999)199827STST\6ST(2000)120(2001)(2002)(20012003)5ST\(2004)(F)(2003)(2006)(2007)(2009)(2009)(2009)(2007)(2005)(2005)(2009)(2007)(2008)(2008)6()(2005)(2009)(2006)(2009)(2004)1.(2005)3STSTFisher7///(2009)t(2006)2003-2006104TFishert-193.3%t-283.7%(2006)FisherX97.2%42.(2004)40ST80ST///()/logit(2005)1998-2000STASTLogitt-2t-33.(2005)(2008)4T(2009)20068T(2009)16ZSPSS16.010Logistic91.110.Carmichael(1972)GeorgeFoster(1986)Ross(1999)()4Whitake(1999).STPTSTSTST[M].2003[J].200111()()()()1.2.123.4.()2.113()()()1.142.3.()154.16.().1122nnzxxxααα=++⋅⋅⋅+12,nααα⋅⋅⋅12,nxxx⋅⋅⋅ZZZZ17.(Logistic)01P(10)()log()ln()1ppitppθ==−ln()1pabxp=+−1epeθθ=+pp0.5p0.5.Probitppabpp0.5p0.5.18.LogisticT-1T-2T-3193.120521()1.()2.()3.()221.2.()——234.11.2.()///1/(1-)///24()·2005-07-1525()()()26()()(2009.27()()()()1.=(4.1)282.=(4.2)3.=(4.3)2=+(4.4)()1.29=(4.5)2.=3.=(4.6)()301.−++-=(4.7)7%~11%2.=(4.8)()1.−=(4.9)FranklinJpbewa,GeorgeT,Friedlob()[M]1999312.−=(4.10)4.1X1X2X9X5X7X8−++-X3X4−X6−X1032CSMARABBASTSTST()()(ST)(ST)ST03-09195STA195ST390STtSTt-1STt-2STt-310SPSS16.033(PrincipalComponentsAnalysis,PCA)Pkn12,,...,nxxx,nn12(,,...,)'nXxxx=uξX11111221221122221122...............kkkkkkkkkkFuxuxuxFuxuxuxFuxuxux=+++=+++=+++(5.1)1.'1iiuu=22212...1(1,2,...,)iiikuuuik+++==2.iFjF(,1,2,3,...,)ijjk≠=k12(,,...,)kFfff=jFXjZxZµσ−=µσ3412,,...,nxxx12,,...,nzxzxzx()KMOKMOSPSS16.05.2()5.2KMOandBartlett'sTestKMOandBartlett'sTestKaiser-Meyer-OlkinMeasureo