基于logistic回归的小微企业信用风险预警

本文以某商业银行化工行业2006年小微企业信贷数据为基础,综合运用因子分析和Logistic回归分析,对小微企业进行客户违约风险预警,并运用2007年的实际违约客户数据对模型风险预警能力进行验证,结果显示模型预警效果较好,这对加强银行的信贷风险管理具有参考意义。     

基于隐马尔可夫模型的物流企业客户违约风险评估

物流企业客户违约及其变化过程具有马尔可夫性。在利用经典算法预测违约损失值的基础上,将违约损失值作为影响因素,借鉴隐马尔可夫理论,构建基于影响因素的物流企业客户违约风险评估模型。通过该模型可以计算得到某一时刻物流企业服务的每个客户的违约风险及其风险变化指数和物流企业面临的平均违约风险及其变化指数。算例计算结果表明,该评估模型能够准确度量客户违约风险,并具有可行性和易操作性。     

机器学习视角下商业银行客户信用风险评估研究

有效把控信贷风险是商业银行稳健运行的关键环节。本文从商业银行客户信贷数据出发,运用非平衡样本处理算法使少数类样本信息得到平衡,并通过机器学习分类器挖掘影响客户违约的重要风险因子,最后构建Logistic模型计算违约概率。研究发现:第一,客户忠诚度是重要因子,忠诚度越高,客户违约概率越低;第二,客户历史信贷数据价值高,是事前风险控制中的重要参考依据;第三,信贷合同特征是影响客户违约的另一重要维度,包括合同期限和合同利率。研究结论可以为银行授信、风险预警和防范违约风险提供理论参考和实践指导。     

考虑ESG表现的企业违约风险预警研究

近年来，企业ESG表现（即在环境保护、社会责任和治理水平方面的表现）在违约风险预警中的作用得到了广泛关注。使用2020年和2021年共417家A股上市公司数据进行回归分析，探究ESG表现与企业违约风险之间的联系，构建加入ESG指标的企业违约风险预警模型，同时进行样本内检验和样本外预测。结果显示，ESG表现与企业违约风险呈显著的负相关关系，较高的ESG得分对应着更低的违约风险，且这种负向影响在非国有企业中更加明显。加入ESG指标能在一定程度上提高预警模型的准确度，有效降低第二类错误。研究结论为考虑ESG因素的企业违约风险研究提供了更有力的参考，并对ESG指标在金融市场上的应用提出了相关建议，对助力我国绿色金融发展有一定意义。     

基于信用消费行为的商业银行零售业务信用风险再评估

商业银行零售业务信用风险评估问题具有维度高、数据更新快等明显特征,在此条件下,利用统计分析、数据挖掘技术建立可靠的动态预警模型具有十分重要的意义。基于数据挖掘的思想,充分利用客户信用消费行为数据,以及粗糙集理论对决策表中条件属性进行约简,并构建基于变精度加权平均粗糙度和基尼指数的决策树算法,依照决策属性值进行客户违约还款预测。实验结果表明,改进后的基于粗糙集与决策树算法的信用卡消费信用风险动态预警模型在准确率、稳定性等方面往往优于基本统计模型及机器学习算法。     

公司信用债违约风险预警与防范研究

我国债券规模位居世界第二,债券市场已成为企业直接融资的主要渠道;同时我国公司信用债违约频发,违约主体几乎涵盖了全部行业,永煤AAA债券违约事件引发各方关注。在此背景下,本文研究了信用债违约风险预警与防范,搭建了债券违约预警模型:一是深入分析了违约原因,提出了"经济下行加剧‘债务-通缩’""流动性分层导致再融资困难""民企互保引发违约风险串联"的观点;二是基于KLR信号分析法,以历史违约主体财报数据为基础构建了上市公司债违约预警模型,抽离出相关指标权重构成预警指标体系,并进行了实证检验;三是基于预警模型,提出加强动态监测、构建债券风险分类管理办法等政策建议。     

基于Lasso-logistic回归的企业集团信用风险研究

通过采用债券违约样本进行实证研究,选取违约主体首次发生信用风险预警信号时点前一年的数据,将多元化的21个风险特征指标加入Lasso-logistic回归模型进行研究,最终选取了11项企业集团信用风险关键预警指标。     

基于监管视角的信用风险预警研究

本文基于监管视角确定了信用风险预警的实现路径,辨析了基于监管视角进行信用风险预警的必要性及其特殊性;以信用风险发生的基本单元即银行客户为研究介质,以监管部门长期监测的银行客户数据、非现场监管数据和经济数据为基础,建立了涵盖客户财务指标、信贷行为、关联担保、区域经济、行业运行的前瞻性指标体系;利用Logistic模型对客户的信用风险进行度量。实证检验结果显示,模型预警效果良好、风险得分前十名的客户,预警抓获率高达65%。为进一步提高预警效率,使之能更好地用于实践,本文通过综合权衡模型的观察面和覆盖面,将风险得分排名前200名的客户确定为预警适宜区,再将2014年6月批次的山东省银行业客户数据代入预警模型,测算当前正常类客户在今后十二个月内变为不良客户的可能性,为信用风险防控提供抓手并赢得宝贵时间。此外,本文还根据预警分析提出监管部门进行信用风险防范、化解的意见建议,以推动银行业机构完善信用风险防范的长效机制。     

个人信贷审批收入支出比信用评价模型探析

银行个人信贷审批人员通常采用收入支出比作为审批标准.本文结合审批经验分析了影响客户收入、支出的因素,将这些因素作为Logistic:回归模型的预测自变量,建立收入支出比客户信用评价模型.应用该模型可得出客户违约可能性的预测值,辅助个人信贷审批人员进行审批决策.     

甘肃商业银行信用风险的研究

本文利用改进的KMV模型,对甘肃省11家上市公司2010年中期财务数据进行分析,得到各个上市公司的违约距离和违约概率,比较了各公司的信用状况,得出甘肃省商业银行信用风险总体较低的结论,最后提出了控制研究甘肃省商业银行信用风险的几点建议。     

Recovery rates,default probabilities,and the credit cycle

In recessions, the number of defaulting firms rises. On top of this, the average amount recovered on the bonds of defaulting firms tends to decrease. This paper proposes an econometric model in which this joint time-variation in default rates and recovery rate distributions is driven by an unobserved Markov chain, which we interpret as the “credit cycle”. This model is shown to fit better than models in which this joint time-variation is driven by observed macroeconomic variables. We use the model to quantitatively assess the importance of allowing for systematic time-variation in recovery rates, which is often ignored in risk management and pricing models.     

Deciphering big data in consumer credit evaluation

This paper examines the impact of large-scale alternative data on predicting consumer delinquency. Using a proprietary double-blinded test from a traditional lender, we find that the big data credit score predicts an individual’s likelihood of defaulting on a loan with 18.4% greater accuracy than the lender’s internal score. Moreover, the impact of the big data credit score is more significant when evaluating borrowers without public credit records. We also provide evidence that big data have the potential to correct financial misreporting.     

非零售类风险暴露信用风险模型的校准和主标尺开发

模型校准是将模型输出结果对应到真实的违约概率。本研究通过一个以违约概率为度量标准的主标尺,映射得到风险等级的过程。该过程引入了所有资产组合风险量化的统一标准。模型的校准和主标尺的设计开发是一个过程中相互联系的两个步骤,该过程受不同条件的约束,是一个多目标优化的问题。本文主要阐述了主标尺开发和模型校准的方法。     

Do defaulting CEOs and directors increase the likelihood of financial distress of the firm?

We hypothesize that the information on a CEO’s and directors’ (board members) past personal payment default entries in public credit data files significantly increases the predictive power of Altman’s (in J Fin 23(4):589–609, 1968) and Ohlson’s (In J Acc Res 18(1):109–131, 1980) distress prediction models. We base our hypothesis on the literature showing that (1) managerial traits such as overconfidence, over-optimism, and the illusion of control affect corporate decisions and that (2) these same personal traits explain personal over-indebtedness and credit defaults. Our results of analyzing the credit data files of more than 100,000 CEOs and directors of the Finnish private limited liability companies support this hypothesis. Our results remain materially unchanged when using the bootstrapping method to assess their significance and when excluding small firms (firm size below the sample median). Collectively, our results imply that creditors should recognize the increased distress risk of firms appointing defaulting CEOs and directors.     

Aggregating sectors in the infectious defaults model

Abstract

How to model the dependence between defaults in a portfolio subject to credit risk is a question of great importance. The infectious default model of Davis and Lo offers a way to model the dependence. Every company defaulting in this model may ‘infect’ another company causing it to default. An unsolved question, however, is how to aggregate independent sectors, since a naive straightforward computation quickly gets cumbersome, even when homogeneous assumptions are made. Here, two algorithms are derived that overcome the computational problem and further make it possible to use different exposures and probabilities of default for each sector. For an ‘outbreak’ of defaults to occur in a sector, at least one company has to default by itself. This fact is used in the derivations of the two algorithms. The first algorithm is derived from the probability generating function of the total credit loss in each sector and the fact that the outbreaks are independent Bernoulli random variables. The second algorithm is an approximation and is based on a Poisson number of outbreaks in each sector. This algorithm is less cumbersome and more numerically stable, but still seems to work well in a realistic setting.     

How Unsecured Credit Policies Influence Mortgage and Unsecured Loan Defaults

Before the global financial crisis, the proportion of households defaulting on the mortgage while remaining current on the unsecured loan was almost the same as the proportion of households current on the mortgage but defaulting on the unsecured loan. After the crisis, the former ratio became higher than the latter. By using a heterogeneous agent model with the mortgage and the unsecured loan, I examine how the order of defaults changed before and after the crisis. I then analyze the impacts of unsecured credit policies on households' mortgage and unsecured loan defaults. My quantitative exercise shows that both default rates can decrease as the cost for unsecured loan defaults increases.     

Constant Proportion Debt Obligations (CPDOs): modeling and risk analysis

Constant Proportion Debt Obligations (CPDOs) are structured credit derivatives that generate high coupon payments by dynamically leveraging a position in an underlying portfolio of investment-grade index default swaps. CPDO coupons and principal notes received high initial credit ratings from the major rating agencies, based on complex models for the joint transition of ratings and spreads for all names in the underlying portfolio. We propose a parsimonious model for analysing the performance of CPDO strategies using a top-down approach that captures the essential risk factors of the CPDO. Our approach allows us to compute default probabilities, loss distributions and other tail risk measures for the CPDO strategy and analyse the dependence of these risk measures on various parameters describing the risk factors. We find that the probability of the CPDO defaulting on its coupon payments can be made arbitrarily small—and thus the credit rating arbitrarily high—by increasing leverage, but the ratings obtained strongly depend on assumptions on the credit environment (high spread or low spread). More importantly, CPDO loss distributions are found to exhibit a wide range of tail risk measures inside a given rating category, suggesting that credit ratings are insufficient performance indicators for such complex leveraged strategies. A worst-case scenario analysis indicates that CPDO strategies have a high exposure to persistent spread-widening scenarios and that CPDO ratings are shown to be quite unstable during the lifetime of the strategy.     

What happens after corporate default? Stylized facts on access to credit

In this paper, we investigate what happens to firms after they default on their bank loans. We approach this question by establishing a set of stylized facts concerning the evolution of corporate default and its resolution, focusing on access to credit after default. Using a unique dataset from Portugal, we observe that half of the corporate default episodes last 5 quarters. Most firms continue to have access to credit immediately after resolving default, though only a minority has access to new loans. Firms have more difficulties in regaining access to credit if they are small, if their default was long and severe, if they borrow from only one bank or if they default with their main lender. Further, half of the defaulting firms record another default in the future. We observe that firms with repeated defaults are, on average, smaller and experience longer and more severe defaults.     

What determines the composition of a firm's cash reserves?

We investigate what determines variation in the composition of the financial assets that constitute corporate cash reserves and how this variation relates to other key liquidity management practices. The degree to which a firm invests its cash reserves in less liquid, longer-maturity financial assets that earn a higher yield is explained by financial constraints, the ability to accurately forecast short-term liquidity needs, and the firm's likelihood of defaulting on its debt. During years when a firm's cash reserves are required to fund increases in investment or operating expenses the firm transfers funds from less liquid to more liquid financial assets. A firm's decisions relating to the composition of its cash reserves interacts with other key liquidity management practices, such as relying on credit lines for liquidity, extending trade credit or using it as a source of financing, and holding large amounts of inventories. Our findings provide insights on an important component of corporate liquidity management decisions.