贝叶斯方法在信用风险度量中的应用研究综述

贝叶斯方法可以有效的处理信用风险度量中常见的数据缺失问题,而且为科学使用专家意见等主观经验提供了有效途径,已被广泛应用于信用风险度量领域。本文从模型构建、估计方法及模型比较三个方面对应用贝叶斯方法度量信用风险的重要文献进行综述,重点关注信用风险的违约相关性和风险蔓延性等最新研究热点,为深入研究信用风险度量问题提供参考,并引起国内风险分析人员对贝叶斯方法的兴趣。     

基于贝叶斯方法的低违约组合违约率估计

现代信用风险建模的核心是估计违约率,违约率估计是否准确将直接影响信用风险建模的质量。在估计违约率的众多文献中,频率法或logistic回归等统计方法的运用非常广泛,此类统计模型的基础是大样本,它客观上需要最低数量或最优数量的违约数据,而低违约组合(LDP)是指只有很少违约数据甚至没有违约数据的组合,如何估计LDP的违约率、反映违约率的非预期波动是一个值得关注的现实问题。本文针对银行贷款LDP缺乏足够历史违约数据的情况,采用贝叶斯方法估计LDP的违约率,并进一步探讨了根据专家判断或者根据同类银行LDP违约数量的历史数据来确定先验分布的方法。在贝叶斯估计中,通过先验分布的设定,不仅可以实现违约率估计的科学性和合理性,而且可以反映违约的非预期波动,有助于银行实施谨慎稳健的风险管理。     

creditrisk~+模型在信贷组合信用风险度量中的应用研究

creditrisk~+模型以泊松分布为理论基础,是一种输入数据较少、计算复杂度较小、便于在银行实际中应用和推广的信贷组合信用风险度量模型.本文在分析国外creditrisk~+模型频带划分缺陷基础上,改用加权平均的频带划分方法,并提出了新的违约率参数的确定办法,利用creditrisk~+模型对大连市商业银行某支行224笔中小公司贷款的非预期损失进行了实证计算.结果发现:creditrisk~+模型可以有效地计量信贷组合的非预期损失且可提高我国商业银行经济资本管理效率.     

地方政府债券信用风险缓释工具定价研究

随着地方政府债券发行规模的扩大,地方政府债务的信用风险日益凸出。本研究以企业债信用风险缓释工具的推出为契机,借鉴结构化模型的思路和KMV模型求解违约概率的逻辑,通过Monte Carlo方法模拟地方政府的违约过程,直接测算地方政府的整体违约概率;结合简约化模型的思路测算地方政府债券的具体违约概率,计算信用风险缓释工具的理论价格,从而构建了地方政府债券信用风险缓释工具的混合定价模型。研究发现,以企业债券为标的测算出的模型理论价格与市场报价基本一致,参数的敏感性检验进一步验证了模型的理论自洽性和实证可靠性。上述结论或将为新《预算法》实施过程中地方政府债务的治理与掌控及中国区域性、系统性金融风险的防范提供新思路。     

违约强度由Lévy从属过程驱动的约化信用风险模型及信用违约互换的定价

本文引入一个约化信用风险模型,其中违约强度定义为从属过程,即非负增Lévy过程.用概率方法得到了违约时间分布的解析表达式.利用该解析表达式,给出了该信用风险模型下的信用违约互换(Credit Default Swaps)的闭形式的定价公式.     

银行信用风险演变的KMV模型分析—以五家中小商业银行为例

利用KMV模型方法,借助预期违约概率（EDF）和违约距离（DD）两个指标分析在我国A股上市的五家中小商业银行的信用风险。着重分析其预期违约概率的变化以及违约距离对股票价格、无风险利率、股权价值波动率等参数的敏感性。结果表明：五家商业银行在2008年前10个月的EDF上升明显,2008年11月EDF开始明显回落。从宁波银行的个案来看,违约距离对无风险利率的敏感性较弱、对股价在较低价位时的敏感性较强,而在较高价位时敏感性较弱,对股权价值波动率的敏感性较强。从违约距离对各参数的敏感性分析结论出发,阐述了稳定并提振我国A股股市的重要性。     

基于KMV模型的我国上市公司信用风险研究

在KM V框架的基础上对股权价值计算方法进行了改进,通过改进后的方法,计算出1999年至2006年各年所有上市公司的违约距离、理论违约率、企业价值、股权价值等指标数据.从分析的结果来看,上市公司规模对信用风险有一定影响,上市公司规模越大,信用风险越小,公司规模越小,信用风险越大.从违约风险的变化情况看,2003—2006年上市公司的违约距离呈下降态势,说明近年来上市公司的违约风险加大.对比沪深300上市公司股改前和股改后信用状况,发现股改前后信用状况有显著不同,股改后上市公司的违约风险变大.通过违约距离的敏感性分析,认为股权价值波动率对违约距离最敏感.     

金融风险度量的建模理论与方法的一些进展及其应用

本文综述了金融风险度量的建模的理论和方法最近的发展。介绍了常用的矩度量和现代风险度量技术,包括在险价值VaR、预期不足ES和期望分位数Expectile等现代风险度量技术和方法,以及复杂风险因素下的非/半参数风险度量方法。违约概率和违约相关性是信用风险度量中的两个基本概念,本文还介绍了信用违约风险中违约概率和违约相关性的常用度量方法。最后,通过一些应用案例介绍如何在金融风险度量中应用现代风险度量技术度量和识别风险。     

金融风险中违约传染效应的研究

市场中的金融资产作为一个整体,往往具有相当复杂的相依风险。本文对信用风险违约的相依性进行研究,提出一种传染效应下违约预报模型,运用加权平均的极大似然方法获得违约强度的估计。并通过对2004年至2008年美国银行业、汽车业和房地产业的数据进行实证分析,得出三大行业间存在着明显的违约风险传染的结论。     

基于不均衡数据的小企业信用风险评价

小企业信用风险评价既是银行风险管理问题,又事关经济社会稳定。针对小企业贷款实践中,违约样本远少于非违约样本、且违约客户误判对银行影响较大的现实,采用不均衡支持向量机对小企业信用风险评价指标进行赋权,进而构建了能有效区分违约客户、非违约客户的评价模型。根据有无特定评价指标、特定评价指标数值变化对贷款小企业违约状态的影响程度赋权;反映了对违约状态影响越大、评价指标权重越大的赋权思路。将违约样本正确识别率、违约样本的准确率与查全率等因素作为支持向量机赋权模型中客户识别率的度量标准,改变了样本数据不均衡所导致的样本总体精度很高、违约样本精度反而不高的现象。研究结果表明:行业景气指数、资本固定化比率、净利润现金含量、恩格尔系数、营业利润率等评价指标对小企业信用风险的影响较大。     

Valuing Credit Default Swap under a double exponential jump di?usion model简

This paper discusses the valuation of the Credit Default Swap based on a jump market, in which the asset price of a firm follows a double exponential jump diffusion process, the value of the debt is driven by a geometric Brownian motion, and the default barrier follows a continuous stochastic process. Using the Gaver-Stehfest algorithm and the non-arbitrage asset pricing theory, we give the default probability of the first passage time, and more, derive the price of the Credit Default Swap.     

Default barrier intensity model for credit risk evaluation

We establish the Default Barrier Intensity (DBI) model, based on the conditional survival probability (also called hazard function barrier), which allows the pricing of credit derivatives with stochastic parameters. Moreover, the DBI is an analytic model which combines the structural and the reduced form approaches. It deals with the impact of the default barrier intensity on the processes around the firm. Using this model we prove the Doob–Meyer decomposition of the default process associated with the random barrier. In this framework, we present the default barrier process as the sum of its compensator (which is a predictable process) and a martingale related to the smallest filtration making the random barrier a stopping time. Furthermore, the DBI as well as the Shifted Square Root Diffusion (SSRD) Alfonsi’s model emphasizes on the dependence between the stochastic default intensity and the interest rate. This model can be useful since it can be easily generalized to all the credit derivatives products such as Collateralized Debt Obligations (CDO) and Credit Default Swaps (CDS).     

Default probabilities in a corporate bank portfolio: A logistic model approach

Analysis and management of credit risk has taken on an increased importance in recent years. New regulations force banks and other financial institutions to make a credible effort to chart and manage the risk associated with their client portfolio. Increased competition in the financial market has also improved the motivation of monitoring the risk/reward relationship on various clients. Modern risk measures such as Credit Risk Capital (CRC) and Risk Adjusted Return On Capital (RAROC) are now well established among banks. One problem in such risk frameworks is to find the expected loss (EL) of the bank portfolio. The EL is based on assumptions regarding the estimated default frequency (EDF) for each client or group of clients. Benchmark models for CRC calculations treat EDFs as exogenous and do not devote much attention to how they can be obtained. This article presents a method of estimating such rates for a retail bank portfolio. The analysis is based on a logistic regression model where financial variables as well as other firm characteristics affect the default probability.     

On the regulator–insurer interaction in a structural model

In this paper, we provide a new insight to the previous work of Briys and de Varenne [E. Briys, F. de Varenne, Life insurance in a contingent claim framework: Pricing and regulatory implications, Geneva Papers on Risk and Insurance Theory 19 (1) (1994) 53–72], Grosen and Jørgensen [A. Grosen, P.L. Jørgensen, Life insurance liabilities at market value: An analysis of insolvency risk, bonus policy, and regulatory intervention rules in a barrier option framework, Journal of Risk and Insurance 69 (1) (2002) 63–91] and Chen and Suchanecki [A. Chen, M. Suchanecki, Default risk, bankruptcy procedures and the market value of life insurance liabilities, Insurance: Mathematics and Economics 40 (2007) 231–255]. We show that the particular risk management strategy followed by the insurance company can significantly change the risk exposure of the company, and that it should thus be taken into account by regulators. We first study how the regulator establishes regulation intervention levels in order to control for instance the default probability of the insurance company. This part of the analysis is based on a constant volatility. Given that the insurance company is informed of regulatory rules, we study how results can be significantly different when the insurance company follows a risk management strategy with non-constant volatilities. We thus highlight some limits of the prior literature and believe that the risk management strategy of the company should be taken into account in the estimation of the risk exposure as well as in that of the market value of liabilities.     

一个多因子信用违约互换定价模型

考虑一个违约互换的多因子的简化模型,通过测度的转化并求解一个多变量Riccati常微分方程而得出了这个模型的解析解.此模型是Duffie-Pan-Singleton(2000)模型的特例,但是这个模型存在解析解,而Duffie-Pan-Singleton模型并不存在解析解.模型的解析解对获得直觉的判断和进行模型的计算都非常有帮助,而且模型的解析解对实证检验也是有帮助的.     

Using equity options to imply credit information

The evolution of credit derivatives has inspired many researchers to investigate the behaviour of credit spreads. Today the growing consensus is that the equity option market provides sufficient information to estimate latent credit parameters. Hull et al. (J. Credit Risk 1(1):3–28, 2005) propose a clever approach to estimate credit spreads from the equity option market. In this paper we first perform a time series analysis to test the conjecture of an existing relationship between credit spreads and implied equity volatility and find strong evidence of a positive relationship. We also propose an extension to Hull et al.’s paper that significantly improves credit spread estimation.     

Credit default swaps: implied ratings versus official ones

Recently, in line with the progressive development of the credit derivatives market, the academic research has begun to explore the relationship between Credit Default Swap market and rating events. In this paper, following a market based approach, we calibrate an Implied Rating model on Credit Default Swap market spreads. The non parametric mapping of Implied Ratings is calibrated on a large data set of Credit Default Swap quotes that includes the years of financial turmoils. This allows also to investigate the existence of possible differences between normal and abnormal market conditions. Unlike other models, the one proposed considers a linear penalty function which allows to evaluate market quotes in a neutral way and to formalize a more computationally efficient programming model. We compare the behaviors of credit rating agencies in different markets (EU and USA) and in different sub-periods, in order to analyze whether Implied Rating changes anticipate or follow the effective rating changes supplied by Fitch Ratings, Moody’s and Standard and Poor’s.     

Analytical pricing of defaultable discrete coupon bonds in unified two-factor model of structural and reduced form models

Pricing formulae for defaultable corporate bonds with discrete coupons (under consideration of the government taxes) in the united two-factor model of structural and reduced form models are provided. The aim of this paper is to generalize the two-factor structural model for defaultable corporate discrete coupon bonds (considered in [1]) into the unified model of structural and reduced form models. In our model the bond holders receive the stochastic coupon (which is the discounted value of a predetermined value at the maturity) at predetermined coupon dates and the face value (debt) and the coupon at the maturity as well as the effect of government taxes which are paid on the proceeds of an investment in bonds is considered. The expected default event occurs when the equity value is not sufficient to pay coupon or debt at the coupon dates or maturity and the unexpected default event can occur at the first jump time of a Poisson process with the given default intensity provided by a step function of time variable. We provide the model and pricing formula for equity value and using it calculate expected default barrier. Then we provide pricing model and formula for defaultable corporate bonds with discrete coupons and consider its duration.     

基于BP神经网络误差函数优化算法的武器装备承制商信用评价模型研究

信用评价是选择武器装备承制商的重要手段.以国标为基础,结合承制商具体情况确定了信用评价指标体系.分析了传统信用评价方法的不足,对经典BP神经网络的误差函数进行优化,优化后的网络模型收敛速度更快,预测精度更高.构建BP神经网络武器装备承制商信用评价模型,仿真实验表明武器装备承制商信用评价可以选用BP神经网络模型.     

企业债券组合的优化问题研究

信用风险已成为金融业面临的最重要风险形式.通过结构模型估计企业债券的未来价格,建立了投资损失函数,并利用CV aR风险度量方法构造了期望收益最大化的债券组合优化模型,最后利用我国债券市场数据进行了实证分析.