【作者】 许文；

【导师】 邓贵仕； 迟国泰；

【作者基本信息】 大连理工大学 ， 管理科学与工程， 2007， 博士

【摘要】 银行贷款决策是商业银行经营管理的核心问题。大多数商业银行的不良资产本质上都是由于资产配置失误而产生。关于商业银行的贷款组合优化模型研究对银行的稳定和发展都具有重大意义。论文基于商业银行的内部风险控制，分别在流动性风险控制、利率风险控制、违约损失控制三方面探讨了商业银行贷款风险决策中的组合优化问题。论文共分六章：第一章绪论部分分析了论文的选题依据、相关研究进展、研究方法、研究的技术路线和研究内容。第二章是基于存量累计风险控制的新增单项贷款组合决策模型研究。第三章是基于copula函数的贷款组合期限结构优化模型研究。第四章是基于违约损失控制的多期资产组合动态优化模型研究。第五章是兼控利率风险和流动性风险的资产负债组合优化模型研究。第六章为结论。论文的主要工作如下：(1)建立了基于存量累计风险控制的新增单项贷款组合决策模型把新增单项贷款与已有贷款的存量组合视为一个新的组合，利用信用风险的扰动项来定量地表示出信贷资产的信用风险，控制在这个组合中的整体风险，建立新、旧贷款统筹考虑的组合贷款优化模型。反映了贷款存量组合累计风险对新增贷款决策的直接影响，改变了现有研究仅仅立足于控制增量贷款组合风险的现状。(2)建立了基于copula函数的贷款组合期限结构优化模型用copula函数拟合短期贷款与中长期贷款的联合分布情况，通过联合分布概率来计算贷款组合的风险价值，选择风险最小时的短期贷款与中长期贷款的比例，由此确定了贷款组合的最优期限结构，对于整体与非正态概率没有苛刻要求的copula函数拟合的联合分布概率反映了贷款组合的真实风险，防范了银行的流动性风险，改变了现有研究大多将资产组合的联合分布假设为多元正态分布，因而低估了资产组合风险的现状。(3)建立了基于违约损失控制的多期资产组合动态优化模型通过逆向递推原理，在考虑下一区段优化配置结果的前提下，控制了本区段单位收益所承担的下偏矩风险，以所有区段全部资产收益最大化为目标，建立基于违约损失控制的多期资产组合动态优化模型。反应了不同区段的单位收益所承担的风险对贷款总体效果的相互影响，在考虑单个区间贷款最优的过程中，优化配给所有区间段的贷款。(4)建立了兼控利率风险和流动性风险的资产负债组合优化模型以贷款利息收益最大化为目标，以利率风险免疫条件和数量结构对称为约束，以线性规划为工具，建立了兼控利率风险和流动性风险的资产负债组合优化模型，解决了资产与负债利率的协调与匹配问题，保护了银行的股东权益，避免和减小了流动性风险，保证了银行资产配给的合法性与合规性。更多还原

【Abstract】 Loan portfolio optimization is the core of management of commercial banks. The mostof non-performing assets are contributed by the assigning mistake. The research on loanportfolio optimization model for commercial banks is most important to realize the basic goaland final goal of management,The paper is based on the inner risk control of commercial banks. The paper separatelybased on liquidity risk control、interest risk control and default loss control discuss theproblem of portfolio optimization model of loan risk decision-making for commercial bank.The paper is divided into six chapters. The first chapter is about the issue selection gist,relative research review, research approach, technical route and research content. The secondchapter is the research on the loan portfolios model considering accumulative risk of theexisting portfolio and increment risk of new loans. The third chapter is the research on loanportfolio term structure optimization model on base of Copula. The fourth chapter is theresearch on multi-stage dynamic optimal model of loan portfolio for commercial banks basedon default loss control. The fifth chapter is the research on optimization model ofAsset-Liability portfolio considering interest risk and liquidity risk. The sixth chapter is theconclusion of the paper. The main works of the paper are shown as follows:(1) The paper sets up the loan portfolio Decision-making Model for IndividualIncremental Loan based on the accumulative risk control of the existing portfolio.The paper takes the Individual Incremental Loan and existing loan asa new loanportfolio. This model uses the perturbation item to represent credit risk; this can control thewhole risk of new portfolio. It sets up loan portfolio optimization model considering new andold loan. The model considers the effect of the accumulative risk of the existing portfolio ofthe old loans and the rational relationship between previous portfolio and the new added. Themodel changes the phenomenon of the present research only controlling the risk ofIncremental Loan portfolio.(2) The paper sets up a loan portfolio term structure optimization model on base ofcopula.Using copula model to the consociation distribution of short term and long term loans,and using VaR method to determine the minimum risk assets portfolio, When the risk is leastchoosing the ratios of short term loan and long term loan, then getting the optimization loanportfolio term structure. The joint distribution of fitting function copula which needn’t requestinteger and abnormal distribution reflects the real risk of loan portfolio, keeps away the liquidity risk of bank, changes the phenomenon of the most present research undervalue theasset portfolio risk because of supposing assets obeys multi-normal distribution.(3) The paper sets up multi-stage dynamic optimal model of asset portfolio based ondefault loss controlBy using Backward Induction Method, under considering the next period asset portfoliooptimization, controlling current period the Downside-risk per profits of bank. Takingmaximizing the whole period asset profits as target, setting up multi-stage dynamic optimalmodel of asset portfolio based on default loss control. This reflects the influence to the wholeloan allocation of different period Downside-risk per profits of bank can bear. By consideringthe single period loan portfolio’s optimal to optimize the whole period loan portfolios.Through considering the single period loan optimization, optimizing the Whole period loanallocation.(4) The paper sets up the optimized asset-liability portfolio model that concurrentlycontrol the interest rate risk and liquidity risk.It takes the maximum interest rate of loans as the target, it takes the interest rate riskimmunity condition and the quantity structure symmetry as restrains, by taking linearprogramming as the tool to set up optimized asset-liability portfolio model that concurrentlycontrol the interest rate risk and liquidity risk.. Through the duration gap and immunityconditions, the paper controls the interest rate risk and protects the equity rights, the modelsolves the harmonization and match problem, and it protects the bank equity against the effectand loss while the market interest rate changing. Through the quantity restrictions of law,code and managing, this paper controls the liquidity risk and ensures the payment ability ofbank; it can ensure the legitimacy and standard of bank assets allocation. It controls theliquidity risk, assure the payment ability and avoid the liquidity crisis.更多还原