盈余质量的定量分析【外文翻译】.doc

1、外文文献翻译译文一、外文原文原文QUANTITATIVEANALYSISOFEARNINGQUALITYEARNINGSQUALITYANALYSISHASLONGBEENKNOWNASTHEQUALITATIVEINVESTMENTMANAGERSBESTDEFENSEAGAINSTLOWQUALITYFINANCIALREPORTINGWHILENOTWIDELYKNOWN,RECENTACADEMICANDPROPRIETARYRESEARCHSHOWSTHATEARNINGSQUALITYMETRICSAREALSOUSEFULINGENERATINGALPHAINTHECONTEXT

2、OFQUANTITATIVEINVESTMENTANDTRADINGSTRATEGIESTHEGRADIENTANALYTICSEARNINGSQUALITYMODELEQMISTHEFIRSTQUANTITATIVEFACTORTHATOBJECTIVELYMEASURESEARNINGSQUALITYACROSSABROADSPECTRUMOFCOMPANIESFORBOTHLONGERTERM312MONTHANDSHORTERTERM13MONTHHOLDINGPERIODSTHEMODELWASDEVELOPEDUSINGRIGOROUSSTATISTICALMETHODSTOENS

3、UREAROBUSTFACTORTHATGENERATESEXCESSRETURNSONASTANDALONEBASISWHILEALSOCAPTURINGAUNIQUEDIMENSIONOFRETURNSNOTCAPTUREDBYOTHERQUANTITATIVEFACTORSTHEENDPRODUCTISAHIGHLYUNIQUEFACTORWITHEXCEPTIONALRETURNSANDLOWCORRELATIONINRELATIONTOOTHERCOMMONLYUSEDFACTORSSUCHASTHOSEDERIVEDFROMESTIMATEREVISIONS,EARNINGSMOM

4、ENTUM,PRICEMOMENTUM,CASHFLOW,CORPORATEINSIDER,GROWTHANDVALUETHEUSACCOUNTINGPROFESSIONANDTHESECURITIESEXCHANGECOMMISSIONSECHAVEWORKEDDILIGENTLYTHROUGHTHEYEARSTODEVELOPTHEMOSTRIGOROUSSYSTEMOFACCOUNTINGPROCEDURESINTHEWORLDNEVERTHELESS,THEREISSTILLASIGNIFICANTGAPAPPROPRIATELYCALLEDTHEEXPECTATIONSGAPBETW

5、EENWHATINVESTORSANDCREDITORSEXPECTANDWHATTHEACCOUNTINGPROFESSIONCANDELIVERTHEEXPECTATIONSGAPEXISTSINPARTBECAUSEPUBLICLYTRADEDCOMPANIESHAVEAGREATDEALOFDISCRETIONINCHOOSINGACCOUNTINGPRINCIPLESANDINMAKINGESTIMATESTHATIMPACTTHEIRREPORTEDFINANCIALRESULTSUNDERGENERALLYACCEPTEDACCOUNTINGPRINCIPLESGAAP,THEA

6、MOUNTOFDISCRETIONTHATACOMPANYHASINPREPARINGFINANCIALSTATEMENTSISCONTROLLEDBYTWOFUNDAMENTALPRINCIPLESOFACCOUNTINGCONSERVATISMANDOBJECTIVITYHOWEVER,INREALITYTHESETWOGUIDINGPRINCIPLESAREOFTENSTRETCHEDTOTHELIMITORIGNOREDWHENCONSERVATISMOROBJECTIVITYISIMPAIRED,EARNINGSQUALITYISCOMPROMISEDWHILE,INTHEORY,A

7、FIRMSACCOUNTINGSTAFFSHOULDEMPLOYPROCEDURESTHATAREOBJECTIVEANDCONSERVATIVE,INPRACTICE,MANAGEMENTHASMANYCOMPETINGMOTIVATIONSTHATDRIVETHEIRCHOICEOFACCOUNTINGPOLICIESANDINFLUENCETHEIRPERIODICESTIMATESBECAUSEOFTHESECOMPETINGMOTIVATIONS,MANYCOMPANIESMANIPULATEACCOUNTINGNUMBERSINORDERTOFACILITATETHEFINANCI

8、ALREPORTINGGOALSESTABLISHEDBYMANAGEMENTINTHISREGARD,VIRTUALLYALLFIRMSWORKINGWITHINTHEBOUNDSOFGAAPUSEMINORACCOUNTING“GIMMICKS”TOPRESENTFINANCIALRESULTSINAPARTICULARLIGHTIE,OVERSTATEORUNDERSTATETHEIRTRUEPROFITABILITYORFINANCIALCONDITIONMICROSTRATEGY,INC,FOREXAMPLE,19992000WASUSINGANAGGRESSIVEREVENUERE

9、COGNITIONPOLICYTHAT,WHILENOTINVIOLATIONOFGAAP,TENDEDTOOVERSTATETHECOMPANYSTRUEPROFITABILITYHOWEVER,ITWASNTUNTILTHESECMANDATEDACHANGEINTHEWAYTHATTECHNOLOGYCOMPANIESACCOUNTFORCONTRACTREVENUETHATTHEMARKETASCERTAINEDTHEEXTENTOFTHEOVERSTATEMENTALTHOUGHACCESSTOEARNINGSQUALITYANALYSISQUALITATIVEORQUANTITAT

10、IVEWOULDHAVEREVEALEDTHEDECEPTIONPRIORTOTHECHANGEINACCOUNTINGRULESAFTERTHESECMANDATEDCHANGE,MICROSTRATEGYWASTWICEFORCEDTORESTATEITSEARNINGSANDITSSHARESFELLOVER98INTHEENSUING12MONTHPERIODWHILEMICROSTRATEGYSTREATMENTOFCONTRACTREVENUESWASVERYAGGRESSIVE,ARGUABLYITWASOPERATINGINAGRAYAREAOFGAAPINMOREEXTREM

11、ECASES,HOWEVER,SOMECOMPANIESGOSOFARASTOCOMMITFRAUDULENTACTSTHATMATERIALLYMISSTATETHEIRFINANCIALSTATEMENTSINWAYSTHATDONOTCONFORMTOGAAPFOREXAMPLE,ENRONUSEDANEXTREMELYAGGRESSIVESCHEMEOFOFFBALANCESHEETFINANCINGINORDERTOHIDEMOUNTINGLOSSESTHEENDRESULTWASARGUABLYONEOFTHEMOSTSPECTACULARFINANCIALREPORTINGDIS

12、ASTERSINHISTORYTHOSEUNLUCKYENOUGHTOHOLDENRONSHARESDURINGTHISPERIODLOSTCLOSETO100OFTHEVALUEOFTHEIRINVESTMENTFINALLY,WHILEINTENTIONALMANIPULATIONOFACCOUNTINGNUMBERSISCOMMON,EARNINGSQUALITYPROBLEMSARENOTALWAYSTHERESULTOFINTENTIONALACTSBYMANAGEMENTFOREXAMPLE,THEQUALITYOFINVENTORIESATLUCENTTECHNOLOGIESAS

13、REPORTEDINTHEIRFIRSTQUARTER10QFILEDMAY2000SUGGESTEDANAPPARENTBACKLOGOFINVENTORYTHATINDICATEDAPOSSIBLESLOWDOWNONTHEHORIZONINALLLIKELIHOOD,THISWASATLEASTINITIALLYACASEOFEARNINGSQUALITYPROBLEMSRESULTINGFROMUNINTENTIONALACTSSLOWSALESNEVERTHELESS,LUCENTSEARNINGSCONTINUEDTODISAPPOINTANDTHESTOCKWASDOWNMORE

14、THAN85INTHEENSUINGTWELVEMONTHSSUBSEQUENTEVIDENCESUGGESTSTHATTHEREMAYALSOHAVEBEENSOMEINTENTIONALMISSTATEMENTSONTHEPARTOFLUCENTMANAGEMENTINORDERTOHIDETHEMAGNITUDEOFTHESALESSLOWDOWNHOWDOCOMPANIESMANIPULATEEARNINGSDESPITETHEEFFORTSOFTHEACCOUNTINGPROFESSIONTOENSUREOBJECTIVITYANDCONSERVATISM,ITISSTILLRELA

15、TIVELYEASYTOMANIPULATEACCOUNTINGNUMBERSTHROUGHEITHERUNETHICALBUTNOTNECESSARILYILLEGALAND/ORFRAUDULENTMEANSTHELISTPRESENTEDBELOWPROVIDESAHIGHLEVELOVERVIEWOFHOWMANAGEMENTCANMANIPULATEACCOUNTINGNUMBERS1RECORDINGFICTITIOUSTRANSACTIONSORAMOUNTS2RECORDINGTRANSACTIONSINCORRECTLY3RECORDINGTRANSACTIONSEARLY4

16、RECORDINGTRANSACTIONSLATE5MISSTATINGPERCENTAGESORAMOUNTSINVOLVEDINATRANSACTION6MISSTATINGTHEAMOUNTSOFASSETSORLIABILITIES7CHANGINGACCOUNTINGMETHODSORESTIMATESFORNOSUBSTANTIVEREASON8USINGRELATEDPARTYTRANSACTIONSTOALTERREPORTEDPROFITSACADEMICRESEARCHONEARNINGSQUALITYANDFUTURERETURNSINADDITIONTOTHEANECD

17、OTALEVIDENCEPROVIDEDBYQUALITATIVEEARNINGSQUALITYSERVICESIE,THOSETHATUSESUBJECTIVEEVALUATIONSOFFINANCIALSTATEMENTSTORENDERANEARNINGSQUALITYGRADE,ACADEMICRESEARCHALSOSUPPORTSTHENOTIONTHATQUANTITATIVEMODELSOFEARNINGSQUALITYCANBEUSEDTOEARNEXCESSRETURNSTHEFOLLOWINGBRIEFREVIEWOFTHEACADEMICLITERATUREHIGHLI

18、GHTSSOMEOFTHEMOSTIMPORTANTFACTORSTHATFORMTHEBASISFORGRADIENTSAPPROACHTOQUANTITATIVELYMODELINGEARNINGSQUALITYANDFORECASTINGRELATEDEXCESSRETURNSTHEVERYFIRSTSTUDIESTOINVESTIGATEISSUESRELATEDTOEARNINGSQUALITYWERECONDUCTEDBYGPETERWILSONOFHARVARDUNIVERSITY1986,1987USINGANEVENTSTUDYMETHODOLOGYWILSONSKEYCON

19、CLUSIONSARETHATOPERATINGCASHFLOWSANDTOTALACCRUALSIE,CHANGESINCURRENTACCRUALSPLUSNONCURRENTACCRUALSAREDIFFERENTIALLYVALUEDANDTHATBOTHAREVALUERELEVANTTHATIS,THEMARKETAPPEARSTOREACTTOTHEDISCLOSUREOFDETAILEDCASHFLOWANDACCRUALDATAVALUERELEVANCEANDTHATCASHFLOWSAREMOREHIGHLYVALUEDTHANACCRUALSDIFFERENTIALVA

20、LUATIONWILSONSBASICFINDINGSAREALSOSUPPORTEDBYANUMBEROFSTUDIESTHATUSEANASSOCIATIONMETHODOLOGY7,INCLUDINGRAYBURN1986,BOWEN,BURGSTAHLERANDDALEY1987,CHARITONANDKATZ1990,LEVANTANDZEROING1990,VICKERS1993,ALI1994,PFEIFFERETAL1998,ANDVICKERS,VICKERSANDBETTIES2000THEFACTTHATTHEMARKETVALUESADOLLAROFCASHFLOWMO

21、RETHANADOLLAROFCURRENTORNONCURRENTACCRUALSIMPLIESTHATHIGHERLEVELSOFACCRUALSAREINDICATIVEOFLOWERQUALITYOFEARNINGSINOTHERWORDS,THEDEGREETOWHICHACOMPANYMUSTRELYONACCRUALSTOBOOSTNETINCOMERESULTSINLOWERQUALITYEARNINGSNEVERTHELESS,ITISPOSSIBLETHATTHEMARKET“SEESTHROUGH”THEDECEPTIONANDAPPROPRIATELYVALUESCOM

22、PANIESBASEDONSOMENOTIONOFBASELINEORSUSTAINABLEEARNINGSHOWEVER,THEFIRSTSTUDIESTOINVESTIGATETHISISSUESLOAN,1996ANDSWANSONANDVICKERS,1997FINDTHAT,CONTRARYTOTHEEFFICIENTMARKETSHYPOTHESIS,DISAGGREGATINGEARNINGSINTOCASHFLOWANDACCRUALCOMPONENTSISUSEFULINIDENTIFYINGSECURITIESTHATARELIKELYTOOUTPERFORMORUNDER

23、PERFORMINTHEFUTURETHUS,THERESULTSOFTHESESTUDIESIMPLYTHATSECURITYPRICESDONOTFULLYREFLECTTHEINFORMATIONCONTAINEDINTHECASHFLOWANDACCRUALCOMPONENTSOFEARNINGSFOLLOWINGINTHEPATHOFSLOAN1996ANDSWANSONANDVICKERS1997,ACADEMICRESEARCHERSARECURRENTLYFOCUSINGONTHEDEVELOPMENTOFSIMPLEEMPIRICALMODELSTHATOBJECTIVELY

24、ASSESSEARNINGSQUALITYINORDERTOPREDICTFUTURERETURNPERFORMANCESEE,FOREXAMPLE,SLOAN,SOLOMONANDTUNA,2001CHAN,CHAN,LEGATEESANDLAKONISHOK,2001ANDPENMANANDZANG,2001TABLE1BELOWSUMMARIZESTHERESULTSOFRECENTACADEMICWORKINGPAPERSTHATFOCUSONTHEPREDICTIVEABILITYOFSIMPLEEARNINGSQUALITYMODELSASSHOWNINTHETABLE,THESE

25、STUDIESFINDTHATCOMPANIESWITHRELATIVELYHIGHLOWLEVELSOFACCRUALSTENDTOUNDERPERFORMOUTPERFORMFORPERIODSOF1236MONTHSAFTERTHEDISCLOSUREOFDETAILEDFINANCIALDATASPECIFICALLY,THERETURNSPREADBETWEENSTOCKSWITHTHEHIGHESTLEVELOFACCRUALSLOWESTEARNINGSQUALITYANDTHELOWESTLEVELOFACCRUALSHIGHESTEARNINGSQUALITYRANGESFR

26、OM88TO217,DEPENDINGONTHEAPPROACHUSEDBYTHEAUTHORSINFORMINGPORTFOLIOSTHEIMPLICATIONISTHATMEASURESOFEARNINGSQUALITYCANBEUSEDINFORMINGPROFITABLEINVESTINGANDTRADINGSTRATEGIESGRADIENTANALYTICSEARNINGSQUALITYMODELTHELATESTACADEMICRESEARCHDEMONSTRATESTHATTHEMARKETDOESNOTFULLYIMPOUNDINFORMATIONABOUTEARNINGSQ

27、UALITYATTHETIMETHATDETAILEDFINANCIALSTATEMENTDATAARERELEASEDTHATIS,ASTATISTICALLYBASEDAPPROACHTOANALYZINGEARNINGSQUALITYCANYIELDPROFITABLEINVESTMENTANDTRADINGSTRATEGIESTHUS,THENEXTSTEPWASTODEVELOPAROBUSTMODELTHATISDESIGNEDTOOPTIMIZETHEEXCESSRETURNSTHATCANBEREALIZEDFROMANEARNINGSQUALITYSTRATEGYTHEGRA

28、DIENTANALYTICSEARNINGSQUALITYMODELEQMHASBEENDEVELOPEDTOACHIEVETHISGOALTHEGRADIENTEQMISTHEFIRSTQUANTITATIVEFACTORTHATMEASURESEARNINGSQUALITYACROSSABROADSPECTRUMOFCOMPANIESTHEEQMPROVIDESTWOWEEKLYSCORESRANGINGFROM1STRONGSELLTO8STRONGBUYFOREACHOFTHETOP5000COMPANIESRANKEDBYMARKETCAPITALIZATIONTHELONGTERM

29、SCOREPROVIDESA18RANKINGBASEDONASTOCKSEXPECTEDFUTUREPERFORMANCEOVERA312MONTHHOLDINGPERIODTHESHORTTERMSCORERANKSEACHSTOCKBASEDONITSEXPECTEDFUTUREPERFORMANCEOVERA13MONTHHOLDINGPERIODINCONTRASTTOCOMPETING,COMMERCIALLYAVAILABLEEARNINGSQUALITYSERVICES,THEOUTPUTFROMTHEEQMISDERIVEDOBJECTIVELYNOTSUBJECTIVELY

30、THROUGHSTATISTICALANALYSISOFACCRUALANDCASHFLOWCOMPONENTSOFEARNINGSTHEMODELWASDEVELOPEDUSINGRIGOROUSSTATISTICALMETHODSTOENSUREAROBUSTFACTORTHATGENERATESEXCESSRETURNSONASTANDALONEBASISWHILEALSOCAPTURINGAUNIQUEDIMENSIONOFRETURNSNOTCAPTUREDBYOTHERQUANTITATIVEFACTORSMORESPECIFICALLY,THEMODELWASCONSTRUCTE

31、DUSINGAMULTIPLEREGRESSIONAPPROACHINCLUDINGREPRESSORSFROMACADEMICRESEARCHANDOUROWNTHEORETICALLYSOUNDPROPRIETARYEARNINGSQUALITYCONSTRUCTSESTIMATEDINPOOLEDTIMESERIES,CROSSSECTIONFOR13SECTORCATEGORIES8EACHSEPARATESECTORMODELINCORPORATESTHEMOSTIMPORTANTDIMENSIONSOFEARNINGSQUALITYFORTHATSEGMENTOFTHEMARKET

32、9WHENCONSIDEREDTOGETHER,THESEDIMENSIONSOR“SUBFACTORS”PROVIDEAMEANSOFRELIABLYRANKINGFIRMSMONOTONICALLYACCORDINGTOBOTHTHEIREXPECTEDMEANANDMEDIANEXCESSRETURNSTHEENDPRODUCTISAHIGHLYUNIQUEFACTORWITHEXCEPTIONALRETURNSANDLOWCORRELATIONINRELATIONTOOTHERCOMMONLYUSEDFACTORSALLGRADIENTMODELSAREDEVELOPEDUSINGAD

33、ISCIPLINEDSCIENTIFICAPPROACHOURAPPROACHCANBECHARACTERIZEDASFOLLOWSVARIABLESPECIFICATIONWEBEGINBYCAREFULLYSPECIFYINGEACHVARIABLETOENSUREPROPERMEASUREMENTANDSCALINGWHENMORETHANONESPECIFICATIONISDEFENSIBLE,WECHOOSETHESIMPLESTSPECIFICATIONONTHETHEORYTHATSIMPLICITYWILLYIELDMOREGENERALIZABLERESULTSMODELIN

34、GTECHNIQUESEACHMODELISESTIMATEDUSINGRELATIVELYSIMPLELINEARANDNONLINEARREGRESSIONTECHNIQUESAGAIN,WEBELIEVETHATSIMPLICITYISTHEKEYTOGENERALIZABILITYSENSITIVITYANALYSESALLMODELSARESUBJECTEDTOSENSITIVITYANALYSESTODETERMINEWHETHERORNOTOURRESULTSAREIMPACTEDBYOUTLIERS,CHANGESINREGIMES,ALTERNATIVEVARIABLESPE

35、CIFICATIONSANDMODELINGTECHNIQUES,ANDSOONPROPERUSEOFINSAMPLEANDOUTOFSAMPLEPERIODSEACHMODELISESTIMATEDUSINGDATAFROMASTRICTINSAMPLEPERIODTHEMODELISTHENTESTEDFORGENERALIZABITY,STABILITY,ANDSOONINANOUTOFSAMPLEPERIODCONTROLFORPOTENTIALTHREATSTOINTERNALANDEXTERNALVALIDITYOURRESEARCHEFFORTSAREDESIGNSTOCONTR

36、OLFORCOMMONTHREATSTOINTERNALANDEXTERNALVALIDITYINFINANCIALENGINEERINGSTUDIESSUCHASSURVIVORSHIPBIAS,HINDSIGHTBIAS,SELECTIONBIAS,ANDSOONCONCLUSIONEARNINGSQUALITYANALYSISISOFTENREGARDEDASTHEQUALITATIVEINVESTMENTMANAGERSBESTDEFENSEAGAINSTLOWQUALITYFINANCIALREPORTINGTHELATESTACADEMICRESEARCHALSODEMONSTRA

37、TESTHATTHEMARKETDOESNOTFULLYIMPOUNDINFORMATIONABOUTEARNINGSQUALITYATTHETIMETHATDETAILEDFINANCIALSTATEMENTDATAARERELEASEDTHATIS,ASTATISTICALLYBASEDAPPROACHTOANALYZINGEARNINGSQUALITYCANYIELDPROFITABLEINVESTMENTANDTRADINGSTRATEGIESTHEGRADIENTEARNINGSQUALITYMODELEQMHASBEENDEVELOPEDTOACHIEVETHISGOALTHEEQ

38、MISTHEFIRSTQUANTITATIVEFACTORTHATOBJECTIVELYMEASURESEARNINGSQUALITYFORTHEPURPOSEOFFORECASTINGFUTURERETURNSTHEMODELWASDEVELOPEDUSINGRIGOROUSSTATISTICALMETHODSTOENSUREAROBUSTFACTORTHATGENERATESEXCESSRETURNSONASTANDALONEBASISWHILEALSOCAPTURINGAUNIQUEDIMENSIONOFRETURNSNOTCAPTUREDBYOTHERQUANTITATIVEFACTO

39、RSTHEENDPRODUCTISAHIGHLYUNIQUEFACTORWITHEXCEPTIONALRETURNSANDLOWCORRELATIONWITHOTHERCOMMONLYUSEDFACTORSTHEEARNINGSQUALITYMODELHASBEENEXTENSIVELYBACKTESTEDACROSSAVARIETYOFSTOCKUNIVERSESANDTIMEPERIODSINORDERTOENSUREOPTIMAL,GENERALIZABLERESULTSTHERESULTSPRESENTEDINTHISWHITEPAPERPROVIDEEXTREMELYSTRONGEV

40、IDENCEONTHEUSEFULNESSOFTHEEQMASSHOWNINTHERESULTSSECTIONOFTHISDOCUMENT,THEMODELPRODUCESHIGHLYSIGNIFICANTEXCESSRETURNS,PERFORMSEXTREMELYWELLBOTHINANDOUTOFSAMPLE,ANDHASALOWCORRELATIONWITHOTHERCOMMONLYUSEDQUANTITATIVEFACTORSAND,INCONTRASTTOCOMPETING,COMMERCIALLYAVAILABLEEARNINGSQUALITYSERVICES,THEOUTPUT

41、FROMTHEEQMISDERIVEDOBJECTIVELYNOTSUBJECTIVELYTHROUGHSTATISTICALANALYSISOFACCRUALANDCASHFLOWCOMPONENTSOFEARNINGSSOURCEGRADIENTANALYTICS,2005“QUANTITATIVEANALYSISOFEARNINGQUALITY”EB/OLHTTP/WWWGRADIENTANALYTICSCOMWEBSITEMARCH,PP2325二、翻译文章译文盈余质量的定量分析盈余质量分析长时间被认为是定性投资经理针对低质量财务报告的最佳防卫。虽然不广为人知，但近期的学术和专有的研究

42、表明，盈余质量指标也对量化投资和贸易战略中用到的阿尔法的产生有帮助。盈余质量的梯度分析模型是第一个量化的因素，客观地计量在广泛的公司为期（312月）盈利的质量和短期（13月）期间举行。该模型是开发、利用严谨的统计方法，以确保一个强大的因素，在一个独立的基础上产生的超额利益，同时也捕获不受其他因素的捕获量的回报的独特维度。最终产品是一种特殊的回报，是相对于其他常用的估计，如修改，盈利势头，价格动能，现金流，企业内部人，增长和价值产生这些因素的相关性较低非常独特的因素。美国的会计界和证券委员会已通过多年的工作努力，发展了世界上最严格的会计程序系统。尽管如此，仍然有很大的差距，（适当地称为“期望差距

43、”），即投资者和债权人的期望与什么是会计专业可以提供的之间的差距。期望差距的存在是因为上市公司在选择夸己原则和作出估计是存在很大的自由裁量权，影响其财务报告的财务结论，按照公认会计原则的自由裁量权的金额，公司已在编制财务报表中规定了两个基本的夸己原则稳健性和客观性。然而，在现实中这两项指导原则，往往是到了极限或忽略。当保守主义或客观性受损，盈余质量受到损害。虽然，在理论上，一个公司聘用会计人员的程序应当是客观的，保守的做法，然而在管理上有许多相互竞争的动机驱使使他们所选择的夸己政策定期估算。由于这些相互竞争的动机，很多公司操纵会计数字，以促进财务报告管理的既定目标。在这方面，几乎所有的企业在会

44、计准则的范围内使用核算小“噱头”并从特定的角度呈报财务结果（即，高估或低估其真实盈利能力或财务状况）。例如，MICROSTRATEGY公司在1999年2000年间采用了激进的收入确认政策，虽然不违反会计准则，但却夸大了公司的真实盈利能力。然后，直到美国证券交易委员会规定的方式中，技术公司的合同收入帐，市场确定的多报（虽然获得收益质量分析的程度变化定性或定量会前透露的欺骗在会计规则的改变）后，美国证券交易委员会规定的变化后，MICROSTRATEGY两次被迫重申其盈利，在随后的12个月期股票下跌了98以上。而MICROSTRATEGY的合同收入的治疗是非常积极的，可以说这是一个公认会计准则经营的

45、灰色地带。在极端情况下，一些公司竟然从事欺诈行为，重大虚报的方式不符合美国通用会计准则的财务报表。例如，安然使用一个极端激进的表外融资方案以隐藏不断增加的损失。最终的结果是有据可查的最壮观的财务报告灾害的历史。那些倒霉的，其投资足以容纳安然公司股票在此期间失去了将近百分之百的价值。最后，会计数字的故意操纵尽管很普遍，盈余质量问题并不总是管理层有意为之的结果。例如，朗讯科技存货的质量（报告于2000年5月第一季度的10Q文件中）显示一个明显表示在地平线上的可能放缓的库存积压。十有八九这是（至少在最初阶段）是盈余质量问题造成的意外行为（慢销售）案件。然而，朗讯的收入继续令人失望，股票在随后十二个月

46、内下跌85以上。（后续的证据表明可能也有一些故意虚报朗讯管理的为了隐藏的销售放缓程度）。公司是如何操纵收益的尽管会计界在努力，以确保客观性和保守主义，但是它仍然是比较容易通过操纵会计数字或不道德的（但不一定是非法的）或欺诈的手段实施的。下面的列表提供了如何高水平综述，管理可以操纵会计数字。1记录虚构交易或数额2不正确记录交易3记录交易早4记录交易晚5虚报百分比或在一个事务中涉及的金额6虚报资产或负债的数额7更改会计处理方法或估计没有实质性的原因8使用关联方交易改变报告的利润盈余质量与未来收益的学术研究除传闻证据的定性收益提供了优质的服务（即那些使用财务报表的主观评价，以使有关收益质量等级），学

47、术研究也支持这样的收益质量的定量模型可以用来赚取超额收益的概念。以下学术文献的简要回顾强调的最重要因素，形成了梯度的定量建模方法的基础盈余质量与预测相关的超额收益部分。探讨收益质量的最早研究者是哈佛大学的G彼得威尔逊（1986年，1987年），使用的研究方法是事件研究法。威尔逊的主要结论是经营性现金流和总应计项目（当期应计项目的变动加上非当期项目）的定价是不一样的，但都是价值相关的。也就是说，市场似乎对现金流细节和应计项目的数据有反应（即价值相关），现金流要比应计项目得到更高的估价（差别估价）。威尔逊的基本结论也支持了研究，利用关联方法学，包括雷伯恩（1986年），鲍文，布鲁克斯大林和戴利（1

48、987年），沙里和卡茨（1990年），黎凡特和泽林（1990），维克瑞（1993）号，阿里（1994年），菲弗等（1998年），和维克瑞，维克瑞和贝蒂斯（2000年）。事实上，市场价值的现金流超过了目前非流动应计费用美元或更多的美元意味着更高水平的预提费用低收益质量的指标。换句话说，在何种程度上依赖于一个公司必须在权责发生制提高质量较低收入纯收入的结果。不过，它可能是市场“看穿”的欺骗和适当的基准值对一些概念或可持续盈利的公司。然而，第一次研究，以探讨这一问题（斯隆，1996年；斯旺森和维克瑞，1997年）发现，与有效市场假说相矛盾的是，将收益分解成现金流量和应计项目在区分可能在将来表现好的与

49、表现不好的证券方面有用处。追随斯隆（1996年）和斯万森和维克瑞（1997）的步伐，学术研究人员目前正在对简单的经验模型，客观地评价收益质量，以预测未来的回报表现。总结了最近的学术工作文件的结果，重点放在预测能力简单的盈利质量模型。这些研究发现，相对高（低）预提费用水平的公司往往会副执行期限为1236个月后，详细的财务资料披露（优于）。具体来说，与应计项目的最高水平（最低收益质量）和预提费用（最高盈利质量）范围的最低水平从88到217，这取决于作者们使用的方法形成的投资组合股票收益之间的传播。其含义是，收益质量的措施，可以形成盈利的投资和贸易战略的方法。梯度分析的盈余质量模型最新的学术研究表明，市场不完全封锁有关在详细财务报表数据发布时的盈余质量信息。这是一种基于统计方法分析盈余质量可以产生有利可图的投资和贸易的战略。因此，下一步是要开发一个健全的模型，设计优化了，可以从一个盈余质量的策略实现超额收益。盈余质量的梯度分析模型（EQM）已发展到实现这一目标。渐变EQM是第一个的定量因素，计量盈余质量于跨广泛的公司。EQM按周提供了两个分数，分别是1强销售到8（强买）为每个最高5000公司的总市值排名。长期提供18排名股票的预期未来业绩超过13个月的持有期。近期成绩排名每个基于持有期的13月的预期未来性能的股票。相对于竞争对手，