1、本科毕业论文（设计）外文翻译原文THEBASISOFCREDITSCORINGONTHEDEFINITIONOFCREDITDEFAULTEVENTSABSTRACTTHECONCRETECIRCUMSTANCESWHICHLEADTOTHEDEFINITIONOFACERTAINCREDITASDEFAULTORNONDEFAULTAREUSUALLYNOTDISCUSSEDINTHELITERATURESINCEWEHAVEACCESSTOALARGEDATASETOFCOMPLETEPAYMENTHISTORIESOFRELATIVELYSHORTTERMEDINSTALLMENTCRE
2、DITS,WECANINVESTIGATEAPOSSIBLESOLUTIONTOTHEPROBLEMOFDEFININGANDDETECTINGDEFAULTEDCREDITSWEPROPOSEADEFINITIONOFCREDITDEFAULTEVENTSWHEREWEADOPTTHEPATTERNSOFPAYMENT,ACOMMONMEASUREFORTHECONTROLOFACCOUNTSRECEIVABLE,TOTHECASEOFINSTALLMENTCREDITSINADDITION,WEDEFINEINDICATORSOFINDIVIDUALPAYMENTPERFORMANCEWH
3、ICHCANBEUSEDFORMONITORINGPAYMENTBEHAVIORONLINE,ONCEACREDITSCORINGSYSTEMHASBEENIMPLEMENTEDINPRACTICECONSEQUENTLY,THEUPDATEOFASCORINGMODELNEEDNOTDEPENDONTHEINFORMATIONOFCLOSEDCREDITS,WHICHBECOMESONLYAVAILABLEATTHEENDOFTHEPAYMENTTERMATTHEEARLIEST1INTRODUCTIONCREDITSCORINGISTHESTANDARDIZEDPROCESSOFANALY
4、ZINGANDCLASSIFYINGTHEPAYMENTBEHAVIOROFINDIVIDUALS,IEPRIVATECONSUMERSORCOMPANIESTHISPROCESSISBASEDONSTATISTICALMETHODSFORESTIMATINGTHEINDIVIDUALPROPENSITYTOSHOWTHEEXPECTEDPAYMENTBEHAVIOR,EGTOSETTLEPAYMENTBYINSTALLMENTSREGULARLYONTHEBASISOFTHISASSESSMENT,CREDITAPPLICANTSAREASSIGNEDTOONEOFTWOOREVENMORE
5、CLASSESTHESECLASSESREPRESENTSUFFICIENTANDINSUFFICIENTPAYMENT,ORCREDITWORTHYANDNOTCREDITWORTHYCUSTOMERS,RESPECTIVELYTHESTATISTICALTECHNIQUESTHATCOVERTHEPROCESSOFMODELINGINDIVIDUALCREDITRISKSAREWIDELYDISCUSSEDINTHELITERATUREEGBYHANDANDHENLEY1997THOMASETAL2002INCONTRAST,WEFINDONLYFEWSTATEMENTSONHOWTHED
6、EPENDENTVARIABLEDEFAULTYES/NOINASCORINGMODELISDEFINEDEVENINSTATISTICALPUBLICATIONS,THISDEFINITIONISALWAYSSAIDTOBEGIVEN,BUTNOTDESCRIBEDNONETHELESS,DEFININGCREDITDEFAULTEVENTSISACRITICALTASKWITHINTHEPROCESSOFMODELINGCREDITRISKASANYSUCHDEFINITIONISNEEDEDTOOPERATIONALIZETHEKEYDEPENDENTVARIABLE,EGCREDITW
7、ORTHINESSITCANBEASSUMEDTHATTHISLACKOFINFORMATIONISDUETOCONFIDENTIALITYREASONSBECAUSETHEDEFINITIONOFCREDITDEFAULTGIVESDIRECTINSIGHTINTOABANKSORCOMPANYSINTERNALCALCULATIONS,ITSCREDITPOLICYANDMARKETINGSTRATEGYINSECT2WEBRIEFLYDESCRIBETHEEMPIRICALDATASETWHICHWEANALYZEINTHECOURSEOFTHISPAPERTOARRIVEATAPOSS
8、IBLEDEFINITIONOFCREDITDEFAULTWEADOPTTHEPATTERNSOFPAYMENT,ACOMMONMEASUREFORTHECONTROLOFACCOUNTSRECEIVABLE,TOTHECASEOFINSTALLMENTCREDITSINSECT3BASEDONAMEASUREOFPROFITABILITYTHATCANBEDERIVEDFROMTHEPAYMENTPATTERNSWECLASSIFYEVENTSOFDEFAULTANDNONDEFAULTINSECT4WEDEFINEINDICATORSOFINDIVIDUALPAYMENTPERFORMAN
9、CEFORTHEMONITORINGOFPAYMENTBEHAVIORANDEVALUATETHEMWITHRESPECTTOTHEIRPOTENTIALTODETECTDEFAULTSONLINE,IEALREADYDURINGTHEPAYMENTPROCESSTHEPAPERCLOSESWITHADISCUSSIONINSECT52DATASETOFINDIVIDUALPAYMENTHISTORIESTOEXEMPLIFYOURPROPOSALSFORDEFININGANDMONITORINGCREDITDEFAULTEVENTS,WEANALYZE33,986INSTALLMENTPUR
10、CHASESOFHOUSEHOLDAPPLIANCES,EACHOFTHEMPAIDOFFBY15REGULARINSTALLMENTSFORTHECOMPANYGRANTINGTHESECREDITS,1329OFTHEFINANCEDAMOUNTSREMAINUNCOLLECTIBLETHEDATASETCONSISTSOFALLCREDITSFORWHICHTHEDUEDATEOFTHEFIRSTRATEOFPAYMENTLAYSBETWEENMARCH01,2004ANDAUGUST31,2004THEREFORE,WEANALYZEACOMPLETECOHORTOFCREDITSFO
11、REACHCREDIT,WEOBSERVETHEPAYMENTHISTORYRUNNINGFROMTHEDUEMONTHOFTHEFIRSTINSTALLMENTUNTILMARCH,2007THEPAYMENTHISTORIESAREGIVENINTHEFORMOFMONTHLYACCOUNTBALANCESTHISIMPLIESANEXACTINSTALLMENTPLANWITHADUEDATEOFEACHINSTALLMENT,BUTWEARENOTGIVENTHEEXACTDATEATWHICHPAYMENTSAREMADECONSEQUENTLY,WEANALYZETHEDATARE
12、LATEDTOEACHINSTALLMENTCREDITONAMONTHLYBASIS3APAYMENTPATTERNAPPROACHTOTHEIDENTIFICATIONOFCREDITDEFAULTEVENTSALOANCONTRACTINTHESPECIALFORMOFAPAYMENTININSTALLMENTALWAYSINVOLVESANINSTALLMENTPLANWHICHDOCUMENTSTHEDUEDATESANDDUEAMOUNTSOFREPAYMENTTHESEDUE,EXPECTEDPAYMENTSCANBECOMPAREDTOTHEACTUALPAYMENTSOFAD
13、EBTORBYMEANSOFTHEINDIVIDUALACCOUNTBALANCESTHEBASICIDEAOFTHEPROPOSEDCLASSIFICATIONISTOBALANCEEXPECTEDANDACTUALPAYMENTSOFDEBTORSATEVERYPOINTINTIMEATWHICHPAYMENTSAREEXPECTEDBYEVALUATINGTHISPATTERNOFPAYMENTSANDTHEPROFITABILITYOFTHEINVOLVEDACCOUNTSWECANDETERMINETHEMAXIMUMPERIODOFDEFICIENTPAYMENTWHICHISAC
14、CEPTABLEFORFINANCIALPURPOSESONTHEDEFINITIONOFCREDITDEFAULTEVENTS31THEPATTERNSOFPAYMENTINTHECONTEXTOFSALESONCREDITTHERECEIVABLEBALANCEPATTERN“ISTHEPROPORTIONOFANYMONTHSSALESTHATREMAINSOUTSTANDINGATTHEENDOFEACHSUBSEQUENTMONTH”JOHNSONANDKALLBERG1986THISPROPORTIONISEXPECTEDTODECAYOVERTHESUBSEQUENTMONTHS
15、THEREFORE,ITISTRACKEDBYSIMPLYFOLLOWINGTHEPERCENTAGESOVERTIMETHECOLLECTIONPATTERNISTHEMIRRORIMAGEOFTHERECEIVABLEBALANCEPATTERN,GIVINGTHECUMULATIVECOLLECTIONSOFTHESUBSEQUENTMONTHSINPERCENTOFCREDITSALESINTHEFOLLOWING,WEDEFINESUITABLEPATTERNSOFPAYMENTFORTHECASEOFINSTALLMENTCREDITSADESCRIPTIONOFTHEORIGIN
16、ALPROCEDURESISGIVENINSTONE1976SUPPOSEWEOBSERVETHECOMPLETEPAYMENTHISTORYOFNINSTALLMENTCREDITSI1,NWITHTOTALFINANCEDAMOUNTSYIWEALSOSUPPOSETHATTHESECREDITSAREPAIDOFFBYANEQUALNUMBERTOFINSTALLMENTS,ANDTHATPAYMENTSAREDUEATREGULARINTERVALSHENCE,PAYMENTSAREOBSERVEDATPOINTSINTIMET1,T,TH,THWITHEVERYTDENOTINGAN
17、OBSERVATIONPOINTOFDUEINSTALLMENTSTHENTDENOTESTHETOTALNUMBEROFINSTALLMENTSANDATTHESAMETIMETHEENDOFTHEAGREEDPAYMENTTERM,ANDHISTHENUMBEROFPOINTSINTIMEH1,HATWHICHWEOBSERVEPAYMENTSAFTERTHEENDOFTHEREGULARPAYMENTTERMHENCE,THDESCRIBETHEENDOFOURCOMPLETEOBSERVATIONPERIODLETYI,TDENOTETHEDUEAMOUNTOFPAYMENTOFCRE
18、DITIATTIMET,WHICHISTHEAGREEDINSTALLMENTATTIMET,ANDLETXI,TDENOTETHERESPECTIVEAMOUNTACTUALLYPAIDATTIMETTHENARETHECUMULATEDPAYMENTSACTUALLYMADEUNTILKWITHK1,THEQUIVALENTLY,THECUMULATEDEXPECTEDPAYMENTSUNTILKAREDENOTEDBYCONSEQUENTLY,ARETHECUMULATEDOUTSTANDINGPAYMENTSATTIMEKOBVIOUSLY,YKXKKATEACHKINTHISRETR
19、OSPECTIVEANALYSISOFPAYMENTSTHECOLLECTIONPATTERNOVERTHPOINTSOFOBSERVATIONCANBECALCULATEDASTHERESPECTIVECUMULATEDPAYMENTSINOFTHEOVERALLFINANCEDAMOUNTFORALLKRESPECTIVELY,THERECEIVABLEBALANCEPATTERNISGIVENBYK/Y,THATARETHERESPECTIVECUMULATEDOUTSTANDINGAMOUNTSINOFTHETOTALEXPECTEDPAYMENT32MEASUREMENTOFPROF
20、ITABILITYOURDEFINITIONOFCREDITDEFAULTEVENTSISBASEDONTHEPROFITABILITYOFACCOUNTSWHICHCANBEMEASUREDUSINGTHEPATTERNSOFPAYMENTDESCRIBEDABOVETOILLUSTRATETHISAPPROACHWEASSUMETHATANACCOUNTISSTILLPROFITABLEATTIMEKIFTHEADDITIONALCOSTSCAUSEDBYDEFICIENTPAYMENTTHATOCCURREDUPTOKARESTRICTLYSMALLERTHANTHEABSOLUTEPR
21、OFITMARGINGENERATEDBYPAYMENTSMADEUPTOKFORMEASURINGTHEPROFITABILITYPKOFACCOUNTSATTIMEK,WEINTRODUCEAWEIGHTAFORTHECUMULATEDPAYMENTSANDAWEIGHTCFORTHECUMULATEDOUTSTANDINGAMOUNTSKHEREAINMAYBEINTERPRETEDASTHEPROFITMARGINANDCMAYDENOTETHERATEOFADDITIONALCOSTSIN,ESPECIALLYINTERESTCHARGES,INVOLVEDINTHECUMULATE
22、DOUTSTANDINGAMOUNTSTHENTHEPROFITABILITYCANBEMEASUREDASHERETHECUMULATEDADDITIONALCOSTSINCORPORATETHESOCALLEDREVOLVINGEFFECTOFCREDITWHICHOCCURSIFDEFICIENTPAYMENTSAREPROTRACTEDOVERACERTAINPERIODLETTDENOTETHEMINIMUMKOFALLOBSERVATIONPOINTSFORWHICHPK0THISMEANS,TISTHEPOINTINTIMEOFTHEPERIODOFDEFICIENTPAYMEN
23、TSATWHICHTHEPERFORMANCEOFCREDITSISNOLONGERACCEPTABLE,WHEREAST1DENOTESTHELASTPOINTINTIMEOFTHEPERIODOFACCEPTABLEPERFORMANCETHISLEADSTOTHEFOLLOWINGCLASSIFICATIONRULECONCERNINGTHEDEFINITIONOFTHEDEFAULTEVENTZCREDITIISASSIGNEDTOTHECLASSOFBADACCOUNTSIFITCONTRIBUTESTOTHEOVERALLLOSS,THATIS,ITSHOWSANOUTSTANDI
24、NGAMOUNTATTOTHERWISECREDITIISASSIGNEDTOTHECLASSOFGOODACCOUNTSWITHDENOTINGTHESUMOFOUTSTANDINGAMOUNTSFORCREDITIATTTHECLASSIFICATIONRULEISFROMAFINANCIALPERSPECTIVE,CLASSIFYINGCREDITSASDEFAULTSBASEDONTASDEFINEDIN5MEANSTHATTHECOSTSOFFINANCINGTHEOUTSTANDINGAMOUNTSEXCEEDTHEPROFITMADEBYTHERECEIVEDPAYMENTSHE
25、NCE,ITMAYBEUSEFULTOGENERALIZETHERIGHTHANDSIDEOF5BYPKWHERE0INTHEABOVEEXAMPLE33APPLICATIONTOTHEEMPIRICALDATASETINTHEEMPIRICALEXAMPLEWEOBSERVETHEINDIVIDUALPAYMENTHISTORIESOFN33,986CREDITSWITHANAGREEDNUMBEROFINSTALLMENTST15ANDANADDITIONALOBSERVATIONPERIODOFH17MONTHSWHEREWEEXPECTACCOUNTSTOBEFINALLYBALANC
26、EDTHEREFORE,THECOMPLETEOBSERVATIONPERIODISTH32MONTHSFIGURE1SHOWSTHEPATTERNSOFPAYMENTINOFTHEOVERALLFINANCEDAMOUNTFOREACHT,THATARETHEEXPECTEDPATTERNYKY,THECOLLECTIONPATTERNXKYANDTHERECEIVABLEBALANCEPATTERNKYTOEXEMPLIFYTHEMEASUREMENTOFPROFITABILITYWECHOOSEA5,C1FORTHEREGULARPAYMENTTERMT1,15ANDA5,C2FORTH
27、EADDITIONALPERIODT16,32AFTERTHEENDOFTHEAGREEDPAYMENTTERMINFIG1THEPROFITABILITYISGIVENINOFTHEEXPECTEDPROFITPKA/AYKTHELEFTVERTICALLINEDENOTESTHEENDOFTHEREGULARPAYMENTTERMTHEVERTICALLINEINTHEMIDDLEDENOTESTHEPOINTINTIMEBETWEENT1ANDTWHICHWEWOULDUSEFORDEFININGDEFAULTEVENTSINTERMSOFPROFITABILITY,IEWHERETHE
28、PROFITABILITYBECOMESNEGATIVETHEDECISIONRULETHATWASFINALLYIMPLEMENTEDBYTHECOMPANYPROVIDINGTHEDATAISILLUSTRATEDBYTHERIGHTVERTICALLINEHERETHEENDOFTHEPERIODOFACCEPTABLEPERFORMANCEISSETTOT127,WHICHISTWELVEMONTHSAFTERTHEENDOFTHEREGULARPAYMENTTERMTHERESULTINGCLASSIFICATIONRULEFORTHEDEFAULTEVENTZISTHENUMBER
29、SOFDETECTEDDEFAULTSANDNONDEFAULTSONBASISOFTHECLASSIFICATIONRULEAREGIVENINTABLE14INDICATORSOFINDIVIDUALPAYMENTPERFORMANCETHEPROPOSEDAPPROACHHASTWOMAINDISADVANTAGESWITHRESPECTTOTHEMONITORINGANDFORECASTINGOFCREDITDEFAULTEVENTSINPRACTICEFIRST,THEANALYSISOFTHEPAYMENTPATTERNSANDTHEPROFITABILITYOFTHEPORTFO
30、LIOOFCUSTOMERSOVERTIMEDOESNOTTELLUSANYTHINGABOUTTHEINDIVIDUALPERFORMANCEOFCREDITSDURINGTHEPERIODOFDEFICIENTPAYMENTTD1,TSECOND,THISANALYSISISBASEDONARETROSPECTIVEANALYSISOFCOMPLETEHISTORIESWHICHAREOBSERVEDFORAVERYLONGPERIOD,EVENAFTERTHEENDOFTHEAGREEDPAYMENTTERMTHISISNOTSUITABLEFORMONITORINGANDFORECAS
31、TINGDEFAULTEVENTSBECAUSESUCHADATABASEISNOTAVAILABLEDURINGTHEACTUALPAYMENTPROCESSWHEREWEOBSERVEPAYMENTSINREALTIMETHEREFORE,WEPROPOSETWOINDICATORSFORMEASURINGTHEINDIVIDUALPAYMENTPERFORMANCETHEBASICIDEAOFTHESEINDICATORSISTOEVALUATETHEDEVELOPMENTOFPAIDAMOUNTSWITHRESPECTTOTHEEXPECTEDAMOUNTSANDTHETIMELINE
32、OFTHEPAYMENTPROCESS41DESCRIPTIONOFINDICATORSTHEFIRSTINDICATORLIK,WHICHWECALLTHEINDIVIDUALLIQUIDITY,RELATESTHECUMULATEDAMOUNTSWHICHAREPAIDUNTILKTOTHERESPECTIVECUMULATEDEXPECTEDAMOUNTSTHEINDIVIDUALLIQUIDITYATTIMEKISTHEREFORETHEPROPORTIONOFDUEPAIDAMOUNTSUNTILKTHESECONDINDICATORPCI,KISCALLEDTHEINDIVIDUA
33、LPAYMENTCAREERANDRELATESTHETOTALSTOCKOFPAIDAMOUNTSTOTHETOTALSTOCKOFEXPECTEDPAYMENTSFIGURE2GIVESTWOHYPOTHETICALEXAMPLESOFINDIVIDUALPAYMENTHISTORIESWHEREATOTALFINANCEDAMOUNTOF600EUROSHASTOBEPAIDOFFBYSIXREGULARINSTALLMENTSOF100EUROSEACHBOTHPAYMENTHISTORIESSHOWDEFICIENTPAYMENTTHECUSTOMERWHOPAIDOFFCREDIT
34、AMISSEDTOPAYATT2,BUTBALANCESTHEACCOUNTIMMEDIATELYATT3THEACCOUNTRELATEDTOCREDITBSTILLSHOWSDEFICIENTPAYMENTATTHEENDOFTHEREGULARPAYMENTTERMFROMTHERESPECTIVEFIGURESOFTHEPERFORMANCEINDICATORS,THEMAINDIFFERENCESBETWEENTHETWOINDICATORSEMERGEWHEREASTHEINDIVIDUALLIQUIDITYEQUALS1AGAINATT3FORCREDITA,THEINDIVID
35、UALPAYMENTPERFORMANCEKEEPSTHEDEFICIENTPAYMENTINMINDFORALONGERTIMEBUTINCONTRASTTOTHEINDIVIDUALLIQUIDITY,THEINDIVIDUALPAYMENTCAREERCANNOTDIRECTLYBEINTERPRETEDINFINANCIALTERMS,EGASPROPORTIONOFDUEPAYMENTSHENCE,ITWOULDBESUITABLETOEVALUATEBOTHINDICATORSINPARALLEL42APPLICATIONTOTHEEMPIRICALDATASETFOREVALUA
36、TINGTHEDEFINEDINDICATORSOFINDIVIDUALPAYMENTPERFORMANCEONBASISOFTHEEMPIRICALDATASETWERESTRICTTHEIRCALCULATIONTOTHEREGULARPAYMENTTERMT15FIGURE3SHOWSTHEBOXPLOTSFORTHEDISTRIBUTIONOFTHEINDIVIDUALLIQUIDITYLEFTANDTHEINDIVIDUALPAYMENTCAREERRIGHTOVERTD15THESEPLOTSEXCLUDEOUTSIDEVALUESTHEDATAAREGROUPEDACCORDIN
37、GTOTHEIRDEFINITIONASDEFAULTANDNONDEFAULTGIVENINTABLE1OBVIOUSLY,BOTHINDICATORSSHOWTHEABILITYTODISTINGUISHBETWEENGOODANDBADACCOUNTS,ANDTHISABILITYISINCREASINGWITHTBUTTHEGROUPSPECIFICDISTRIBUTIONSOFBOTHINDICATORSALSOSHOWANAREAOFINTERSECTION,EVENFORLARGETTOEXEMPLIFYTHEEARLYDETECTIONOFDEFAULTEVENTSWECHOO
38、SELI,K1/6,THATIS,WECALCULATETHENUMBEROFDEFAULTSANDNONDEFAULTSSHOWINGANINDIVIDUALLIQUIDITYOFLI,K1/6FORK1,TUSINGTHISCUTOFFVALUE,WEWOULDDETECT5479OFTHE5,382CREDITSDEFAULTSALREADYATTD6FORTD12THISPROPORTIONINCREASESTO5762,ANDTO6193ATT15ATTHESAMETIME,THEPROPORTIONOFNONDEFAULTSDETECTEDBYTHISCUTOFFVALUEFORT
39、HEINDIVIDUALLIQUIDITYDECREASESFROM156ATT6TO066ATT12ANDTO050ATT15ALTHOUGHESPECIALLYTHEMISCLASSIFICATIONERRORISRELATIVELYLOW,ITHASTOBEREMEMBEREDTHATTHISCLASSIFICATIONDIRECTLYDEPENDSONTHEDEFINITIONOFDEFAULTIN8ANDTHEUNDERLYINGCOSTRATIOWHENMEASURINGTHEPROFITABILITYFURTHERMORE,THEANALYSISISBASEDONEMPIRICA
40、L,HISTORICALDATAOFCREDITHISTORIES,ANDTHISIMPLIESTHATTHECOMPANYTOOKACTIONTOCOLLECTOUTSTANDINGAMOUNTSITHASTOBEASSUMEDTHATTHEANALYSISISBIASEDASTHESEACTIONSDOORDONOTHAVEANEFFECTONTHEINDIVIDUALPAYMENTBEHAVIOROFCUSTOMERSSOURCEALEXANDRASCHWARZANDGERHARDARMINGER2010“THEBASISOFCREDITSCORINGONTHEDEFINITIONOFC
41、REDITDEFAULTEVENTS“STUDIESINCLASSIFICATIONDATAANALYSISANDKNOWLEDGEORGANIZATIONPART3PP595602译文信用评分法的基础信用违约事件的定义摘要定义导致信用违约行为的具体情况。通常，在文献中都没有讨论如何确定信用是否违约。由于我们拥有关于短期分期付款完整付款记录的大量数据，我们可以研究一个可能的方法来解决关于如何定义和鉴定拖欠账款信用这个问题。在采用付款模式时提出信用违约事件的定义，在分期信用事件中用一个共同的测定方法来控制应收账款。此外，一旦在线信用评分系统在实践中得到了落实，我们定义的个人支付性能指标就可用于监
42、控付款行为。因此，更新的得分模型不需要依赖封闭的信用，只适用于早期付款期结束时。1、介绍信用评分是对个别支付行为进行分析和分类的标准化进程，换而言之，针对私人消费者或公司。这个过程是基于数理统计等方法，对个人估算表明预计的倾向，例如安排定期的分期付款来进行支付。在此基础上，对评估信用的申请人指定给一个、两个（甚至更多）的等级。这些等级分别代表支付充足或不足的情况，或者代表信誉卓著或信誉不好的客户。统计技术涉及到建模过程，个人信用风险在文献中被广泛地讨论（如汉德和亨利1997年，托马斯孙俐2002年）。相反，我们发现只有少数的报表依据可变的债务是否是有效模型而进行定义的判断。即使在统计出版物中，
43、这一定义也总被提到要进行考虑，但是并没有详细地说明。尽管如此，定义信用违约事件是建立信用危机模型过程中一个关键的任务，可以运用一些关键的相关变量，例如商誉，它被认为是由于保密的原因导致缺乏信息。信用违约的定义由银行或公司内部的计算，及其信用政策和市场策略提供直观的见解。在第二部分，我们将简短地描述对这份文件进行分析的过程中所得到的经验数据。为分期支付信用，尽可能作出一个关于我们所采取的支付模式下的信用违约的定义，一个控制应收帐款的共同标准。在第三部分，基于一种衡量盈利的方法，推导出一个可以分类出违约和不违约支付的模型。在第四部分，我们为了监测付款行为和评估他人有关他们潜在的违约底线，而定义个人
44、支付表象的指标，即已经在付款过程中。这篇文章以第五部分的讨论作为结束。2、个人支付历史的数据确定为了举例说明我们为定义和监控信用违约事件的建议，我们对33,986个分期购买家用电器的案例进行分析，他们中的每一个人分15次定期付清账款。为了让公司同意提供这些信用，资金数量的1329不能收回。确定的数据来自于2004年3月1日和2004年8月31日之间的所有应收账款日期最初支付比率的信用。因此，我们分析了一组完整的信用。为了了解每个信用，我们观察从第一部分到2007年3月的付款历史。支付历史以每月帐户余额的形式给出。这意味着有每个分期至到期日的精确分期付款计划，但我们不能提供确切的支付日期。因此，
45、我们在一个月的基础上进行分期支付信用的数据分析。3、定义信用违约事件的支付方式的方法一份租借合同在特殊形式的支付分期通常会涉及到分期付款的文档截止期限和应付的数量。这些应付账款，预期的支付额通常比作债务人依靠个人账户余额的实际支付款。其基本思想是平衡预期和实际支付的分类。债务人在任何时候进行支付都在意料之中。通过评价该付款模式和有关账户的盈利能力，确定帐户在金融方面能够被接受的可拖欠账款最大期限，确定信用违约事件。31付款模式在信用应收平衡模式的销售背景下“任何一个月的比例销售额在随后每个月月末保持出色的表现”（约翰逊和盖博，1986年）。这一比例被期望在未来的几个月里下降。因此，它就是简单的
46、随着时间推移而变化的百分比。收集模式是应收账款平衡模式的镜像，提供所收集的随后几个月的信用销售百分比的累积数据。在下面，我们将定义分期信用合适的支付模式。用石材描述了初始的过程（1976）。如果我们观察完整的N的付款历史，分期付款信用I1,N和总资产数额。我们也认为这些信用等于分期T的数量和定期间隔的应付账款。因此，在时间点上观察到的支付额T1,T,TH,TH，每一个T代表分期应付的观察点。然后T代表分期总数和同时约定的付款期的结束，H代表时间点的数量H1,H，在定期付款期限结束后我们所观察到的支付。因此，TH描述了我们完整的观察期。设YI,T为在时间T内，被批准的分期付款的到期信用支付金额，
47、设XI,T为时间T内各段时间支付的实际数额。则实际上，累计报酬直到K，K1，TH。相同的，到K时段的累计预计报酬到表示为因此，是时间K内累计拖欠的工程款。显然，在每一个K内，YKXKK。在这个国际收支平衡收集模式的回顾性分析中，在T这个观察点可计算各自累积报酬在整体金额中的百分比数量其中在所有的时间K中XK/Y。相应的，应收账款余额模式由K/Y决定，其表示分别累计未付款项在总的预收款项里的百分比。32利润的测量我们关于信用违约事件的定义是基于可以被付款模式描述和衡量的账户盈利状况。为了说明这方法，我们假定一个帐户在K时段中是盈利的，如果因为发生K比营业收益小，就会出现没有付款引起费用增加的情况
48、。用于测量账户在K时段内的盈利状况PK，我们介绍了一个为累计支付的变量A和累计未完成数量K的变重C。A在这里可以被解释为边际收益，C可以表示额外费用的比率，尤其是利息费用，参与累计未完成的金额。则盈利能力表示为在这里，当如果拖欠报酬超过一定期限，累积额外费用就会纳入所谓的信用循环效应。设T表示PK0的所有观测点的最小值K。这意味着，T是在拖欠付款期间信用的履行不接受延长的时间点，而T1表示在可接受的履行期间最后的时间点。这导致了以下涉及到违约事件Z的分类规则如果信用I会增加整体亏损，它就被归于坏账类，也就是说，它显示了在时间T的欠款数额。否则信用I就归为良好帐目类表示信用I在时间T的未付款项总
49、和的分类规则为从财政角度来看，信用分类作为基于时间T的拖欠事项定义为（5）除应收账款利润之外未付账款数量的财务费用。因此，在上面的例子中它可能由右侧（5）PK当0概括。33应用实证的数据集在实证的例子中，我们观察到约定的分期付款数量T15和另一个H17个月的我们期望帐户最终平衡的观察期的N33,986信用的个人支付历史。因此，完整的观察期是TH32个月。图1显示了每一个T的整体融资资金百分比的支付模式，预期模式YK/Y，集合模式XK/Y和应收平衡模式K/Y。距离证明盈利的可测量性，我们选择定期付款期T1,15中A5，C1和在约定付款期结束后的额外周期T16,32中A5，C2。在图1中给出预期利润PK/AYK的盈利能力。左边垂线表示最后分期付款的期限。垂线表示T1和T之间的时间点，我们将用来盈利期间的信用事件，例如盈利能力在哪里为负。终于执行的决策由公司在右垂线上提供合适的数据来说明。在约定付款期后的12个月内，可接受性能的时期末设T127。由此产生信用违约事件Z的的分类规则是图1给出在分类规则上违约事件和非违约事件的数量4、个人支付模式的指标这个被提到的方法在实践中关于追踪和就监测与预测信用违约事件时有两个主要的缺点。首先，支付模式的分析和客户随着时间推移的盈利能力不能告诉我们在未完成付款
