信用评分模型的拒绝偏差与Heckit纠正

信用评分模型在构建过程中,样本数据通常仅是那些贷款申请被接受,贷款违约与否信息能被观测到的数据,这一样本数据缺陷导致模型在应用中出现被称为拒绝偏差的参数估计偏误,影响了模型的预测准确度。本文的研究采用微观计量经济学中的Heckit方法,借助商业银行的住房按揭贷款微观数据对信用评分模型的拒绝偏差问题进行实证研究。研究结果表明住房按揭贷款信用评分模型存在拒绝偏差,而经过Hecikt方法纠正的信用评分模型能有效地提高模型的预测能力。     

基于Logistic模型的个人信用评分体系研究

本文通过收集的样本客户信用报告资料拟合了Logistic模型,并设计了客户基本分计算方法、客户平时信用表现计分标准、客户最终得分计算方法、客户信用等级评判标准等.本文设计的个人信用评分体系既利用了客户基本信息判断其成为好客户的概率(向前预测),又充分考虑了其以往信用表现(历史信息),在评价客户信用状况时做到了前期工作客观公正.     

我国小企业信用评分系统的开发策略

为了缓解借贷双方信息不对称问题,防范和化解借款人较高的逆向选择和道德风险,促进其盈利水平和核心竞争能力的提高,商业银行将信用评分技术引入小企业货款业务中,使用数据对小企业信用风险进行准确度量,准确快速地识别"好客户"和"坏客户".目前小企业信用评分模型已广泛应用干欧美等发达国家中小企业授信业务上,并在大量授信实践中丰富了信用评分的内容,使该项技术日趋成熟.     

我国小企业信用评分系统的开发策略

为了缓解借贷双方信息不对称问题，防范和化解借款人较高的逆向选择和道德风险，促进其盈利水平和核心竞争能力的提高，商业银行将信用评分技术引入小企业贷款业务中，使用数据对小企业信用风险进行准确度量，准确快速地识别“好客户”和“坏客户”。目前小企业信用评分模型已广泛应用于欧美等发达国家中小企业授信业务上，并在大量授信实践中丰富了信用评分的内容，使该项技术日趋成熟。     

信用评分模型开发中的抽样技术

信用评分是运用数据挖掘技术对已知客户的信息进行分析，建立能预测未来客户信用表现的模型。数据准备是评分模型开发过程中非常重要的步骤，数据质量的好坏直接决定了模型的成败。由于银行内部的数据量非常庞大，为了使分析更加有效率，需要对数据进行抽样。因此，如何进行抽样，如何保证样本能够充分代表总体就非常重要。根据信用评分模型的开发经验以及数据挖掘中的抽样理论，现提出如下建立评分模型时应用的抽样技术以及注意事项。     

信用链优化与小企业融资脱困研究

在借贷资本运动的二重支付和二重回流这个信用链中，当前小企业融资难主要是其在一重支付和二重回流这两个环节受阻所致。需要通过增强小企业抗风险能力、改革金融业经营管理、提高小企业信用能力、增强小企业信用意识等措施优化小企业的信用链，帮助小企业走出融资困境。     

个人信用评估组合模型的构建——基于决策树—神经网络的研究

本文以德国某商业银行的真实客户信用数据为样本,将决策树方法应用于个人信用指标的筛选过程中,并与BP神经网络模型相结合构建成一个两阶段组合模型。研究表明,基于决策树—神经网络构建的个人信用评估组合模型对于测试样本的分类预测精度高于单一BP神经网络模型的分类预测精度。组合模型对于测试样本的总正确率平均值为75.45%,高出单一BP神经网络模型的总正确率近3个百分点。基于信息熵增益率分类原理的最优决策树挑选指标方法能合理去除非重要属性指标的干扰,使真正有效的属性指标输入神经网络主模型,提高模型分类预测的精准度。     

珠联璧合:基于机器学习的网络借贷信用评分卡模型研究

近年来,P2P网络平台"爆雷"事件频发,除平台恶意"携款跑路"外,绝大多数平台倒闭的原因在于部分借款人违约而导致平台现金流不足,进而出现提现困难、严重挤兑。本文借助人人贷网络借贷平台的数据,利用广义交叉验证、随机森林等机器学习算法对借贷客户特征变量进行筛选,并建立融入证据权重的Logistic信用评分卡模型,根据客户贷前所提交的个人信息对客户进行信用评分,进而预测客户借款到期时的违约情况。实证结果表明,该模型的预测准确率明显高于样本期间的平台评级,且模型通过了稳定性检验。这一研究结论对P2P网络借贷行业的健康发展与个人信用体系的构建具有重要意义,同时对第三方评级机构的个人信用评估也有一定的参考价值。     

债券市场的评级信息能改善股票市场信息环境吗?——来自分析师预测的证据

本文研究发现,信用评级增加了分析师预测的公共信息,同时没有减少分析师预测的私有信息,即信用评级信息能够改善股票市场信息环境。具体地,本文利用2006-2016年上市公司数据实证发现,信用评级能显著提高分析师预测精度、降低分析师预测分歧度与乐观偏差,且这种效应在信息不对称程度高、低能力分析师跟踪的企业以及外资参股的评级机构中更加显著,这说明信用评级向分析师提供了新的信息。进一步地,信用评级只增加了分析师预测的公共信息,对私有信息没有显著影响,同时也不影响分析师的调研行为,上述结论表明信用评级信息披露并不影响分析师拥有私有信息的优势,且信用评级机构与分析师之间存在互补而非替代关系。因此,金融监管应该加强对评级机构尽职调研的要求,充分发挥信用评级机构的信息中介功能。     

我国小企业贷款难题及其化解

在我国,由于金融市场发展相对滞后、社会信用环境不佳等原因,小企业贷款难的问题十分突出。一、小企业贷款难的原因分析1.信息不对称因素。一是人才因素制约。许多小企业规模小,会计人员力量比较薄弱,一些企业甚至没有专门的财务管理人员,财务信息的可信性差。二是小企业数量众多,信用缺失问题严重。长期以来财务报表不规范,给工商、税务、银行几套报表数据都不尽相同。三是小企业贷款缺乏可靠的历史数据,也没有可信的外部评级等,     

Variable reduction,sample selection bias and bank retail credit scoring

This paper investigates the effect of including the customer loan approval process to the estimation of loan performance and explores the influence of sample selection bias in predicting the probability of default. The bootstrap variable reduction technique is applied to reduce the variable dimension for a large data-set drawn from a major UK retail bank. The results show a statistically significant correlation between the loan approval and performance processes. We further demonstrate an economically significant improvement in forecasting performance when taking into account sample selection bias. We conclude that financial institutions can obtain benefits by correcting for sample selection bias in their credit scoring models.     

个人信用评分关键技术研究的新进展

从系统论的角度总结了个人信用评分发展的前沿问题,从数据预处理、指标体系筛选、以及模型设计三个方面对个人信用评分关键技术的最新研究成果进行了细致分类和综合比较,从而指出个人信用评分研究中存在的难点以及未来发展方向。     

Modelling small‐business credit scoring by using logistic regression,neural networks and decision trees

Previous research on credit scoring that used statistical and intelligent methods was mostly focused on commercial and consumer lending. The main purpose of this paper is to extract important features for credit scoring in small‐business lending on a dataset with specific transitional economic conditions using a relatively small dataset. To do this, we compare the accuracy of the best models extracted by different methodologies, such as logistic regression, neural networks (NNs), and CART decision trees. Four different NN algorithms are tested, including backpropagation, radial basis function network, probabilistic and learning vector quantization, by using the forward nonlinear variable selection strategy. Although the test of differences in proportion and McNemar's test do not show a statistically significant difference in the models tested, the probabilistic NN model produces the highest hit rate and the lowest type I error. According to the measures of association, the best NN model also shows the highest degree of association with the data, and it yields the lowest total relative cost of misclassification for all scenarios examined. The best model extracts a set of important features for small‐business credit scoring for the observed sample, emphasizing credit programme characteristics, as well as entrepreneur's personal and business characteristics as the most important ones. Copyright © 2005 John Wiley & Sons, Ltd.     

Voluntary Disclosure of Sales and the Extent of Trade Credit in Small Private Companies

We examine the association between voluntary financial disclosure and the amount of obtained trade credit in a sample of small private Belgian companies. We argue that voluntary disclosure can help small private companies in mitigating information asymmetries that arise between the company and their suppliers. Using a propensity score matching procedure to control for selection bias, we find that voluntary financial disclosure by small and private companies is positively related to the level of trade credit. This is in line with the traditional view that asymmetric or incomplete information restricts access to external funds.     

Credit Scoring and the Availability of Small Business Credit in Low‐ and Moderate‐Income Areas

This paper estimates that credit scoring is associated with about a $3,900 increase in small business lending per sample banking organization, per low‐ and moderate‐income (LMI) area served, and this effect is roughly equivalent to that estimated for higher‐income areas. For our sample, this corresponds to a $536 million increase in small business credit in LMI areas in 1997 than otherwise would have been the case. This effect appears to be driven by increased out‐of‐market lending by banking organizations, as in‐market lending generally declines. Overall, it does not appear that credit scoring has a disparate impact on LMI areas.     

The Surprising Use of Credit Scoring in Small Business Lending by Community Banks and the Attendant Effects on Credit Availability, Risk, and Profitability

The literature has documented a positive relationship between the use of credit scoring for small business loans and small business credit availability, broadly defined. However, this literature is hampered by the fact that all of the studies are based on a single 1998 survey of the very largest U.S. banking organizations. This paper addresses a number of deficiencies in the extant literature by employing data from a new survey of the use of credit scoring in small business lending, primarily by community banks. The survey evidence suggests that the use of credit scores in small business lending by community banks is surprisingly widespread. Moreover, the scores employed tend to be the consumer credit scores of the small business owners, rather than the more encompassing small business credit scores that include data on the firms as well as on the owners. Our empirical analysis suggests that credit scoring is associated with an initial increase in small business lending activity that moderates over time and no change in the quality of the loan portfolio. Supplementary analysis suggests that the use of credit scores for small business lending has a negative initial effect on community bank profitability that moderates over time.     

The impact of bank consolidation on small business credit availability

This paper examines how banking consolidation has affected small businesses credit. Using the Survey of Small Business Finances, the empirical model examines how credit supply to small firms responds to larger banks, and whether the non-bank supply of credit has offset decreases in credit from banks. Using an empirical model to correct for sample selection, large banks are found to lower the probability of obtaining credit for small businesses, and this lower probability is not offset by increased total loans. Non-bank institutions are found to make up much, but not all, of the decrease.     

Economic benefit of powerful credit scoring

We study the economic benefits from using credit scoring models. We contribute to the literature by relating the discriminatory power of a credit scoring model to the optimal credit decision. Given the receiver operating characteristic (ROC) curve, we derive (a) the profit-maximizing cutoff and (b) the pricing curve. Using these two concepts and a mixture thereof, we study a stylized loan market model with banks differing in the quality of their credit scoring model. Even for small quality differences, the variation in profitability among lenders is large and economically significant. We end our analysis by quantifying the impact on profits when information leaks from a competitor’s scoring model into the market.     

业主信用评分法:小微企业贷款技术的创新

我国小微企业当前的信贷环境,除了企业自身高成本和高风险的因素之外,还与不完善的小微企业贷款定价机制等有关。小微企业的信贷价值和业主的信贷价值是密切相关的,从小微企业主个人的有关信用历史,就可较准确地预期小微企业未来的还款表现,这是业主信用评分法的基本思想。本文总结了业主信用评分法在美国的发展与应用,通过分析我国小微企业贷款信用评分的现状与问题,提出应借鉴业主信用评分法,促进我国小微企业贷款的创新。     

行为评分模型在个人信用评估应用中的实证研究

本文将Logistic模型和马尔科夫链模型相结合,在Logistic模型的基础上综合考虑客户行为状态的变化,将其加入信用评估模型中,得到优于单一运用Logistic模型的结果,据此得到的动态信用评分,为商业银行信贷决策及客户关系管理决策提供更有力的依据。