摘要
贷款组合优化决策是商业银行在综合考虑其贷款风险和贷款收益的前提下,从众多贷款申请企业中选取一组最优贷款对象的过程,它是商业银行信贷管理的核心问题。商业银行通过调整资金的运用,改善其资产配置水平,从而实现资产的流动性、盈利性和安全性。受全球金融危机的影响,各行业生产经营困难加剧,以至于导致商业银行合理协调资金配置实现“三性”平衡的难度增大,商业银行亟待探讨新的资金管理及风险控制手段,因此基于风险控制的贷款组合优化决策模型研究具有重要的现实意义。
本论文共五章进行阐述。第一部分是绪论,主要介绍本文的选题意义及国内外资产负债优化管理理论的研究现状;第二部分是模型原理阐述,为模型构建奠定理论基础;第三部分是模型构建,构建目标函数及大量约束条件;第四部分为应用实例与结果分析,通过实证及对比分析证明该模型的合理性和有效性;第五部分是结论。
本文的主要研究工作如下:
(1)揭示了基于下偏度的损失控制原理。根据偏度原理,运用贷款组合下偏度来衡量收益率的概率密度函数的“左尾”的偏斜方向和偏斜程度。大量研究表明,贷款组合的收益率并非服从正态分布,而是呈现“尖峰厚尾”的形状,而收益率概率分布的“左尾”表示实际收益率低于预期收益率的概率,是真正的贷款组合风险,因此,用下偏度控制风险,更符合投资者的心理,也更符合实际。
(2)构建了基于下偏度最小化的贷款组合优化模型。依据下偏度损失控制原理,结合数学规划方法,以投资者风险厌恶为出发点,以组合风险价值VaR为约束条件,构建基于下偏度最小化的贷款组合优化模型,反映贷款组合的“左尾”风险,减少极值事件发生对银行造成的重大损失。
本文的特色与创新表现为两个方面:一是下偏度不要求贷款收益服从正态分布,能够很好地反映收益率分布的“左尾”,降低商业银行贷款组合发生重大损失的概率;二是下偏度可以真实反映贷款的本质,符合投资者的心理,并且可以反映多笔贷款之间的相关联系,解决现有模型的解析能力不足的问题。
Loan portfolio optimization decision is a process that the commercial banks select a group of optimal loan objects through many loan application enterprises based on the comprehensive consideration of the loan risk and loan earnings. It is the core problem of the commercial bank credit management. The commercial bank improve its asset allocation level by adjusting funds apply, so as to realize the liquidity, profitability and safety of assets ("three properties").The global financial crisis made the business grow difficultily, so that increased the difficulty of keeping "three-properties" for commercial banks. Commercial banks must explore a new method of capital management and risk control immediately. So, Loan optimization model based on risk controlling has important practical significance.
The paper is divided into five chapters. The first part is the introduction, mainly introduces the significance and the development process and research status of portfolio theory. The second part expatiates the basic theory of the model,so can provide the theoretical basis for the model construction. The third part is the construction of optimal model of loan portfolio, including target function and a large number of constraints.The forth part is the practical analysis, proving that the model is reasonable and effective through empirical analysis and comparison. The last part is the conclusion.
The main results of the paper are as follows;
(1) The paper revealed the theory of loss control based on downside skewness. According to the skewness principle, the paper measured the deflection direction and deflection degree of yield's probability density function by using downside skewness of the loan portfolio. A mass of researches indicated that loan portfolios yields was not normal distribution, but the distribution of "peak thick tail".The "left end" of yields probability distribution is the real risk, and shows the probability that the actual yields less than the expected rate. So, controlling the risk with the downside skewness satisfies the investor's psychological and be more actual.
(2) The paper constructed optimization model of downside skewness minimum on loans portfolio. Controlling the risk based on the principle downside skewness, combining with the mathematical programming method, investors'risk aversion as a starting point, value at risk as constraint of assets'risk, then the optimization model of downside skewness minimum on loans portfolio is set up. The model can reflect the "left rail" of the loan's yield primly, and reduce the probability of serious losses of commercial bank.
The contribution of this paper lies on two aspects. Firstly, the downside skewness don't demand that the loan's yield is normal distribution, reflect the "left rail" of the loan's yield primely, and reduce the probability of serious losses of commercial bank. Secondly, Measuring loans risk by the downside skewness could meet the investor's psychology, reflect the relationship among loans and resolve the problem that the existing model analytical ability of existing model is poor.
引文
[1]Perter Rose商业银行管理[M].北京:机械工业出版社,2001:232-244.
[2]中国金融稳定报告(2011)[EB].[2011,06,14].http://www.pbc.gov.cn/publish/jinrongwendingju/370/2011/2011061417173485084674 1/20110614171734850846741_.html
[3]Markos A. Katsoulakis, Andrew J. Majda, Dionisio G. Vlachos. Coarse-grained stochastic processes and Monte Carlo simulations in lattice systems. Journal of Computational Physics,2003,186:250-178.
[4]陈鹏,吴冲锋,李为冰.信用风险度量和管理方法研究[J].管理工程学报,2002,6(6)485-490.
[5]Markowitz, H. Portfolio Selection:Efficient Diversification of Investments. New York:John Wiley&Son, Inc.1959.
[6]Gordon Alexander and Alexander Baptista. A VaR-Constrained Mean-Variance Model: Implications for Portfolio Selection and the Basle Capital Accord. Working paper, University of Minnesota.2001(7).
[7]Kaplanski, Guy. Traditional beta,downside risk beta and market risk premiums.The Quarterly Review of Economics and Finance Volume:44, Issue:5,2004(12):636-653.
[8]W. F. Sharpe. A Simiplified Modei for Portfolio Analysis. Management Science,1963, 10(2):277-292.
[9]Jensen M C. Risk:the Pricing of Capital Assets and the Evaluation of Investment Portfolio. Journal of Business.1969.4
[10]Ross S A. The Arbitrage Theory of Capital Asset Pricing. Journal of Economic Theory. 1976(12):343-346.
[11]Jarrow, R, D. Lando and S. Turnbull. A Markov Model for the Term Structure of Credit Risk Spreads. Review of Finanical Studies,1997,10:481-523.
[12]Merton,R. On the pricing of corporate debt:The risk structure of interest rates. Journal of Finance,1974, (28):449-470.
[13]Merton, R. On the Pricing of Corporate debt:the Risk Structure of Interest Rates [J]. Journal of Financial Economics.1977(5):241-249.
[14]Michael B. Gordy. a Compatative Anatomy of Credit Risk Models. Journal of Banking &Finance 2000(24):119-149.
[15]Wilson, T. Portfolio Credit Risk I.Risk 10(9),1987,9.
[16]Markowitz, H. Portfolio Selection[J]. Journal of Finance.1952,7(1):77-91.
[17]Li Duan, Ng Wan-Lung. Optimal Dynamic Portfolio Selection:Multiperiod Mean-VaRiance Formulation[J]. Mathematical Finance.2000,10(3):387-406.
[18]姜灵敏.基于综合风险收益的贷款组合优化决策[J].数理统计与理.2006,24(6):84-88.
[19]迟国泰,洪忠诚,赵志宏.基于行业组合的贷款总体风险优化决策模型[J].管理学报.2007,4(4):398-403.
[20]Monica B, Loriana P.Value-at-Risk:a Multivariate Switching Regime Approach[J]. Journal of Emprical Finance.2000,7(5):531-554.
[21]Rockafellar, R. T., Uryasev, S.:Optimization of Conditional Value-at-Risk.J. Risk 2000,2,21-41.
[22]刘小茂、李楚霖、王建华.风险资产组合的均值—CVAR有效前沿(Ⅰ)[J].管理工程学报.2003,17(1):28-34.
[23]郭战琴,周宗放.基于VAR约束的商业银行贷款组合多目标决策[J].系统工程理论方法应用.2005,14(2):149-152.
[24]迟国泰,王际科,齐菲.基于CVaR风险度量和VaR风险控制的贷款组合优化模型[J].预测.2009,28(2):47-52.
[25]P. A Samuelson. The Fundamental Approximation Theorem of Portfolio Analysis in Terms of Means, Variances and Higher Moments[J]. The Review of Economic Studies.1970, 37 (4):537-542.
[26]P. Chunhachinda, K. Dandapani, S. Hamid, and A. J. Prakash. Protfolio Selection and Skewness:Evidence from International Stock Markets. Journal of Banking & Finance. 1997,21:143-167.
[27]A. J. Prakash, C. H. Chang, and T. E. Pactwa, Selecting a Portfolio with Skewness:Recent Evidence from US, European and Latin American Equity Markets. Journal of Banking and Finance.2003,27:1375-1390.
[28]荣喜民,崔红岩.基于偏度的多期组合投资调整模型[J].运筹与管理,2005,(14):104-108.
[29]迟国泰,迟枫,闫达文.贷款组合的“均值—方差—偏度”三因素优化模型[J].运筹与管理.2009.18(4):98-111.
[30]Altman,E. L. Measruing Corporate Bond Momality&Performance[J]. Journal of Finance 1989,44 (4):909-922.
[31]Altman, E. L,and D. L. Kao. The Impolications of Coporate Bond Ratings Drift [J]. Financial Analysts Journal,1992, (6):64-75.
[32]Standard &Poor's, Rating Performance 1996:Stability and Transition[A]. Special Report[R].New York:Standard&Poor's,1997.
[33]Cariy, L. Moody's Rating Migration and Credit Quality Correlation:1920-1960[A]. Mood's Special Report[R]. July,1997.
[34]David Lando,Torben M.S. Analyzing Rating Transitions and Rating Drift with Continuous Observations[J]. Journal of Banking & Finance,2002,26 (2-3):423-444.
[35]顾乾屏,陈顺发,刘炜等.信用迁移矩阵特性实证分析[J].石家庄经济学院学报,2007(30):24-29.
[36]洪忠诚,迟国泰,许文。基于信用风险迁移条件风险价值最小化的贷款组合优化模型[J].系统管理学报,2009(18):276-283.
[37]Kusy M I,Ziemba W T. A Bank Asset and Liability Management Model[J]. Oper Res,1986, 34(3):356-376.
[38]Mulvey J M, Ruszczynski A. A New Scenario Decomposition Method for Large-scale Stochastic Optimization[J]. Oper Res,1994,43:477-490.
[39]Leonard C. Maclean, Rafael Sanegre, Yonggan Zhao, William T. Ziemba. Capital Growth with Security[J].Journal of Economic Dynamics & Control,2004,28(2):937-954.
[40]王慧,陈华友,王自强.模糊随机环境下带个人偏好的投资组合决策模型[J].运筹与管理,2008 17(2):131-135.
[41]何超林,孟卫东.基于具有不确定性随机扩散模型的动态资产组合选择[J].预测,2011,30(2):62-65.
[42]Gjerde O, Semmen K. Risk-based Capital Requirements and Bank Portfolio Risk[J]. Journal of Banking and Finance,1995,19(7):1159-1173.
[43]TOKAT Y, RACHEV S T, SCHWARTZ E S. The Stable Non-Gaussian Asset Allocation:A Comparison with the Classical Gaussian Approach[J]. Journal of Economic Dynamics and Control,2003,27(6):937-969.
[44]姚天祥,刘思峰.基于一曲收益率的模糊投资组合优化模型[J].统计与决策,2007(19):47-48.
[45]洪忠诚,迟国泰,吴灏文.基于存量与增量全部组合收益最大的贷款优化决策模型[J].运筹与管理,2008,17(1):114-121.
[46]Morgan, J B and T L Galculation of an Efficient Frantier for a Commercial Loan Portfolio[J]. Journal of Portfolio Management,2003, (1):28-32.
[47]Rasmussen K M, Clausen J. Mortgage Loan Portfolio Optimaization Using Multi-stage Stochastic Programming[J]. Journal of Economic Dynamics and Control,2006,30(1):1-25.
[48]迟国泰,洪忠诚,许文.基于存量与增量全部组合风险最小的贷款优化决策模型[J].管理科学,2007,20(6):84-90.
[49]Consigli G,, Dempster M A H. Dynamic Stochastic Programming for Asset-Liability Management[J]. Annals of Operations Research,1998,81:131-161.
[50]Shing, G, and Nagasawa, H. Interactive decision system in stochastic multiobjective portfolio selection[J]. International Journal of Production Economics,1999,60-61:187-193.
[51]Maclean L C, Saneger R, Zhao Y, et al. Capital Growth with Security[J], Journal of Economic Dynamics and Control,2004,28(5):937-954.
[52]秦学志,迟国泰.多准则多目标信贷策略的动态规划方法.中国管理科学,2000,8:18-24.
[53]姜灵敏.基于综合风险收益的贷款组合优化决策[J].数理统计与管理,2006,24(6):84-88.
[54]马志卫,刘应宗.基于蒙特卡洛模拟的贷款组合优化决策方法[J].管理科学,2006,19(3):66-70.
[55]Uryasev, Stan, Theiler, Ursula A,Serraino, Gaia. Risk-return optimization with different risk-aggregation strategies[J]. Journal of Risk Finance,2010, 11(1):129-146.
[56]王春峰,屠新曙,厉斌.效用函数意义下投资组合有效选择问题的研究[J].中国管理科学,2002,10(2):15-19.
[57]刘艳萍,王婷婷,迟国泰.基于风险价值约束的贷款组合效用最大化优化模型[J].系统管理学报,2009,18(2):121-129.
[58]袁子甲,李仲飞.参数不确定和效用最大化下的动态投资组合选择[J].中国管理科学,2010,18(5):1-6.
[59]常浩,荣喜民.随机参数和随机资金流环境下基于二次效用函数的投资组合优化[J].应用数学学报,2011,34(4):703-711.
[60]余婧.均值—方差—近似偏度投资组合模型和实证分析[J].运筹学学报.2010.14(1):106-114.
[61]许文,迟国泰,杨万武.基于违约损失控制的商业银行多期资产组合动态优化模型[J].管理学报.2010,7(4):585-594.
[62]Fishburn P C. Foundations of Risk Measurement, I. Risk as a Probable Loss[J]. Management Science.1984,30(4):396-406.
[63]Stambaugh, F., Risk and value at risk, European Management Journal.1996(14):612-621.
[64]王春峰.金融市场风险管理[M].天津:天津大学出版社.2001:481-489.
[65]吴珊珊.基于风险价值约束的银行贷款组合优化模型[D].大连理工大学,2005.
[66]郭家华.VAR方法在我国银行风险管理中的应用[J].统计与决策.2010,17(1):172-173.
[67]顾乾屏,陈顺发,刘炜,杜家悌.信用迁移矩阵特性实证分析[J].石家庄经济学院学报,2007,30(2):24-30.
[68]Caouette J B. Managing Credit Risk the Next Great Financial Challenge[M]. New York John Wiley& Sons,Inc,1998:274-280.
[69]迟国泰,董贺超,刘艳萍.基于信用风险迁移的组合收益与组合风险计量模型[J].科研管理.2009,30(2):139-149.
[70]迟国泰,奚扬,姜大治,林建华.基于VaR约束的银行资产负债管理优化模型[J].大连理工大学学报.2002,42(6):750-758.
[71]迟国泰,孙秀艳,董贺超.基于信用等级修正和半绝对离差风险的银行资产组合优化模型[J].系统工程理论与实践.2006,8:1-16.