信用风险评估研究

i基于贝叶斯方法的商业银行信用风险评估研究摘要风险是亘古不变的研究课题,自从信用和金融诞生以来，风险就伴随其左右。每当人们自认为掌握了其中的道理的时候，它就以“经济危机”的形式给人类教训,2008年度的金融危机就是这种教训的典型代表。危机之后，各国政府便开始寻找其背后原因并实施应对措施。在2010年，G-20通过了新的巴塞尔协议即《巴塞尔协议III》，巴塞尔协议III它不仅加强了微观层面的监管，更加注重了宏观审慎的监管。巴塞尔协议III的产生反应了过去对金融风险监管的低效率事实，也说明了信用风险问题从来没有被彻彻底底的解决过。信用与违约这对孪生兄弟，在经历人类数百年的研究之后，依然充满了神秘，依然有待深入研究。另外，我国“十二五”规划纲要中也明确强调，在新的五年里面我国要全面推动金融改革开放，构建多元、服务高效、监管审慎、风险可控的金融体系。商业银行在我国金融体系中占有举足轻重的地位，其信用风险状况关系到我国金融业整体的安全，研究商业银行信用风险具有重要现实意义。综上本文确定研究的对象：信用风险。伴随计算机技术的发展，贝叶斯统计也得到了快速发展，它逐渐成为统计学发展的前沿阵地,这是促使作者选择贝叶斯方法来研究信用风险问题的最初动力。另外，R、SAS、OpenBUGS、WinBUGS等软件的出现也使贝叶斯模型的实践成为可能,由此作者确定了完整的研究题目：基于贝叶斯方法的商业银行信用风险评估研究。本文采用理论与实证相结合的方法，在搜集参考了大量资料的基础上，从商业银行交易对手（上市企业）的角度出发，研究不同贝叶斯信用风险度量模型在商业银行风险管理上的应用问题。研究过程中，我们选择商业银行的交易对手作为研究对象，一类是存在潜在风险的ST企业，另一类是相对安全的非ST企业；通过对企业的17个财务指标作因子分析处理，最终选择能够反映指标大部分信息的6个主因子作为构建贝叶斯模型的影响变量；运用挑选出的敏感性主因子指标分别构建贝叶斯判别分析、贝叶斯Logistic回归模型、贝叶斯Poisson回归模型、贝叶斯二元分位数回归模型。实证结果表明，贝叶斯方法在信用风险度量上不仅具有传统风险度量模型的准确性，还拥有更好的稳健性。关键词：信用风险；贝叶斯；判别分析；广义线性混合模型；二元分位数回归iiTheresearchoncreditriskassessmentofcommercialbankbasedonBayesianmethodsAbstractRiskwasaneverlastingtopicanditwasassociatedwithcredit.Andwheneverpeoplethoughtthattheyknewitstruth,itwouldpunishhumansintheformofeconomiccrisis,andfinancialcrisisin2008wasatypicalexample.Afterthecrisis,thegovernmentsaroundtheworldbegantolookforreasonsandmakesomemeasures,andIn2010,theG-20unanimouslyadoptedtheBaselIII.ThenewbaselIIInotonlystrengthenedthemicro-levelsupervision,butalsopaidmoreattentiontomacro-prudentialsupervision.Anditalsoreactedthatthesupervisionhaslowefficiencyinthepast,thecreditriskissueshadneverbeenthoroughlysolved.Thetwinsofcreditandriskwerestillfullofmysteryandtheyalsoneededtoberesearchedinanexperienceofhundredsofyears.Inaddition,theoutlineofthe12thFive-YearPlanofChinaalsoexplicitlystressedthatweshouldpromotethefinancialreformingandopening,andbuildacompletefinancialsystemwhichwouldhavemultivariablecompositions,efficientserviceandprudentialsupervision.CommercialbankoccupiedapivotalpositioninChinesefinancialsystem,anditsriskwouldberelatedtothesecurityofthewholefinancialindustry,soitwasofpracticalusetodoresearchonthecreditriskofcommercialbank.Consequentlywemadethe“creditrisk”tobeourresearchproject.Withthedevelopmentofcomputertechnology,thebayesianstatisticsalsotookagreatstepforward,andithadbecometheforefrontofstatistics.Thisphenomenonmotivatedmetostudycreditriskproblemusingbayesianstatistics.AndtheemergenceofstatisticalsoftwaresuchasR,SAS,OpenBUGSandWinBUGSmadetheempiricalanalysisinthispapertobepossible.Lastlytheauthormadethedecisionthattheresearchtopicwas“TheresearchoncreditriskassessmentofcommercialbankbasedonBayesianmethods”.Thispaperresearchedthecommercialbanks’riskmanagementproblemsbybuildingdifferentBayesiancreditriskmeasurementmodels,adoptingthecombinationoftheoryandempiricalmethodsandcollectingalargeamountofdataandreferences,andthearticleperspectivewasonthebasisofcounterpartiesofcommercialBanks(listedcompanies).Ontheresearch,wedividedthecounterpartiesofcommercialBanksintotwocategories,onehadapotentialriskmarkedastheSTenterprises,theotheronewhichwastherelativesafetyflaggedasnonSTiiienterprises.Withthehelpoffactoranalysisofseventeenfinancialindicators,wefinallychosesixsensitivemainfactorsastheindependentvariableinthebayesianmodels,andthesefactorscouldreflectmostoftheinformationoftheindicators.Thentheauthorbuiltbayesiandiscriminantanalysis,bayesianLogisticregressionmodel,bayesianPoissonregressionmodelandbayesianbinaryquantileregressionmodel.Theempiricalresultsshowedthatthebayesianmethodhadnotonlythesameaccuracyasthetraditionalriskmeasurementmodel,butalsohasbetterrobustness.Keywords:creditrisk;bayes;discriminantanalysis;generalizedlinearmixedmodels;binaryquantileregressioniv目录论文总页数：55页第一章绪论...................................................................................................................11.1课题背景.............................................................................................................................11.2本课题研究的意义.............................................................................................................11.3研究思路和研究框架.........................................................................................................21.4国内外研究现状................................................................................................................41.4.1国外研究状况..........................................................................................................41.4.2国内研究状况..........................................................................................................51.4.3贝叶斯方法在信用风险中运用的研究概况..........................................................6第二章基于贝叶斯方法的信用风险评估理论基础...................................................72.1信用风险评估理论基础.....................................................................................................72.1.1信用风险定义及诱因..............................................................................................72.1.2信用风险的特征......................................................................