资本账户开放对浮动汇率制度福利的非线性效应:以新兴市场经济体为例

通过构建一个带有随机冲击的开放经济模型,采用比较静态分析与动态分析相结合的方法,探讨了新兴市场经济体资本账户开放对浮动汇率制度福利变化的影响.结果显示,资本账户开放度对浮动汇率制度的福利变化具有显著的非线性效应.浮动汇率制度"吸收器"功能的有效发挥不仅取决于一国资本账户开放度偏离其最优资本账户开放度的位置,还受本国资本账户开放度变化量的影响.得到我国资本账户开放进程中人民币汇率制度改革的启示是:我国仍应在坚持可控、渐进的原则下,逐步推进人民币汇率制度改革与资本账户开放,并注意两者之间的搭配与协调.     

平衡损失下有限总体回归系数的线性可容许预测

本文研究了平衡损失函数下正态总体和非正态总体中有限回归系数的可容许预测.利用统计决策理论,获得了非正态总体中齐次线性预测为可容许预测的充分必要条件和在正态总体中齐次线性预测在一切预测类中可容许性的充要条件,推广了二次损失下的若干相关结果.     

基于交易流水的信用卡套现交易及商户识别

信用卡套现是一种威胁正常金融秩序的风险行为。有效识别具有套现风险的商户及其风险交易,对信用卡风控具有重要意义。传统的信用卡风险识别方法需要先积累大量的标注数据,对持卡人拥有充分的先验信息。而信用卡标注数据的稀缺大大限制了传统方法的应用。本文充分挖掘海量交易流水数据,提出基于无监督学习的套现交易及风险商户识别方法。一方面,该方法无需关于信用卡的标注数据或先验信息,能够以数据驱动的方式过滤行为异常的套现交易及风险商户,具有更广泛的应用前景。另一方面,该方法综合商户的交易金额属性,及商户与消费者之间的关联关系,构建出一系列可解释性强的套现风险指标,为风控管理提供直观的指导参考。基于某第三方支付平台实际数据的实证分析表明,本文方法能够有效区分具有不同行为表现、不同风险等级的商户群体,为实际的套现交易识别提供可靠的决策支持。     

天然水油藏水侵计算实用求解方法

对于具有边水或者底水能量的油藏,进行储量计算或动态预测时,必须考虑天然水域对油气藏开发动态的影响,准确评价油藏天然能量大小.针对定态水侵、准定态水侵、修正定态水侵和非定态水侵四种水侵类型,基于物质平衡原理,给出了实用求解方法.对于边底水油藏水侵能量计算和开发指标预测具有一定意义.     

基于PT-MA理论的煤矿安全群体行为演化博弈分析

将前景理论(PT)和心理账户(MA)理论相结合引入到矿工群体、监管群体的演化博弈决策过程中,构建了效价账户和成本账户前景收益感知矩阵,得到不同情况下演化系统所达到的均衡状态,发现当实施安全行为和监管行为与实施对立行为的价值感知收益偏差呈同化正向收敛时,安全行为和监管行为的演化趋于理想状态。但在现实状况下,群体行为演化并不能达到理想的结果,原因如下:(1)安全行为成本、监管行为成本较高,安全行为效价、监管行为效价较低;(2)效价感知函数参照点高,成本感知函数参照点低;(3)矿工、监管群体面对效价收益时更倾向于风险厌恶,面对成本损失时更倾向于风险追求;(4)群体决策更倾向于低估高概率事件(事故概率、监察概率、举报概率)。最后构建Matlab算法,进行演化博弈数值模拟,验证了调节各参数对行为演化的影响作用。本文为煤矿安全监管提供了新的思路,同时对其他领域的安全监管提供了借鉴价值。     

我国货币政策运行状态的量化分析与预测——基于非平稳离散选择模型和SVM方法比较

本文利用非平稳离散选择(NSD)模型,在两种时间刻度下,对我国央行货币政策操作中调整法定存款准备金率和定期利率的动态行为进行量化分析和预测,并与支持向量机(SVM)的预测结果进行比较.结果表明,核心经济变量及其纵向相对水平变化对央行货币政策调控决策具有显著且较优的解释能力.根据样本外模型预测结果,本文认为以月度为单位对央行制定执行货币政策行为进行分析预测比以季度为单位更为合适.虽然SVM模型整体样本外预测能力优于NSD模型,但NSD核心差分变量模型对央行上下调整政策行为具有较好的预测能力。本文结论对央行的货币政策调整决策行为具有一定的解释能力,有助于市场主体衡量经济运行状态,及时把握央行的货币政策操作动向.     

心理账户、损失厌恶与行为资产配置实证研究

基于前景理论和三参照点理论,建立了单心理账户和三心理账户下的线性损失厌恶行为投资组合模型,并利用中证基金指数数据构建了不同市场状态下的行为投资组合,实证研究不同损失厌恶系数、不同参照点、不同心理账户资金配置条件下模型的最优资产配置策略和投资组合绩效,研究发现线性损失厌恶模型更关注下侧损失,损失厌恶系数影响资产配置,注重安全性的投资者偏好低风险资产,而寻求实现抱负水平的投资者更偏好高收益资产。     

正态分布下任意秩有限总体中的Minimax预测

对于任意秩有限总体,在二次损失下,有关文献已给出了线性可预测变量在齐次线性预测类中的唯一线性Minimax预测.本文在正态假设下,证明了这个线性Minimax预测也是线性可预测变量在一切预测类中的唯一Minimax预测.     

考虑感知成本的共享单车用户借还车行为意向研究

为分析奖励措施下共享单车用户的借还车行为意向,弥补现有用户行为意向无法定量描述的不足,文章基于计划行为理论,提出了考虑不同感知成本的用户借还车行为意向的定量估计方法.首先从用户行为意向的影响因素出发,添加成本感知变量建立用户行为意向模型,利用预调查获取共享单车用户的个体特征和出行特征,通过正交实验法设计了关于出行目的、奖励金额与额外步行时间的出行场景问卷;然后利用收集的用户行为意向数据,基于出行效用理论建立用户行为选择概率模型,通过SPSS软件进行二项logistic回归求解模型参数,实现用户借还车行为意向的定量估计.研究结果表明:无论是借车还是还车,用户早高峰出行均具有更高的时间敏感性,当额外步行时间大于8 min时,奖励金额以及规范停车区的设置对用户借还车行为的影响不明显;而当额外步行时间小于5 min时,奖励金额和规范停车区的设置能较好的影响用户的借还车行为.     

基于双子支持向量机的信用卡流失分析

银行信用卡业务属于高收益、高风险的业务,如何实现对信用卡的客户流失控制是发卡银行迫切需要解决的问题.目前,随着银行积累了大量的数据,并建立了数据仓库,使得采用数据挖掘技术来实现信用卡客户流失分析成为了可能.利用双子支持向量机,基于某商业银行的信用卡数据,建立了信用卡流失分析模型,实验结果证明了方法的有效性.     

Samsung card lending model

Samsung Card Lending Model (SCLM) analyzes cash flow in individual accounts and measures the level of company-wide risk. Serving as a risk and portfolio management model in the consumer lending business, the main features of SCLM are as follows. Default ratios such as intrinsic balance default probability and annual default ratio are computed using the past, present, and future cash flows of accounts. The provision is shown as the total sum of write-offs. The size of capital required is determined by default probability distribution. The price for new accounts is quoted based on cash flow simulations reflecting future business environments. SCLM has shown good performance in Samsung card consumer lending business since the Korean credit card crisis of 2003.     

Intensity models and transition probabilities for credit card loan delinquencies

We estimate the probability of delinquency and default for a sample of credit card loans using intensity models, via semi-parametric multiplicative hazard models with time-varying covariates. It is the first time these models, previously applied for the estimation of rating transitions, are used on retail loans. Four states are defined in this non-homogenous Markov chain: up-to-date, one month in arrears, two months in arrears, and default; where transitions between states are affected by individual characteristics of the debtor at application and their repayment behaviour since. These intensity estimations allow for insights into the factors that affect movements towards (and recovery from) delinquency, and into default (or not). Results indicate that different types of debtors behave differently while in different states. The probabilities estimated for each type of transition are then used to make out-of-sample predictions over a specified period of time.     

国有商业银行企业不良贷款的主因子分析

论文首先在银行企业不良贷款成因的研究成果进行回顾梳理基础上,从探究现行发展环境下,影响企业不良贷款的主要原因的角度,通过对国有商业银行贷款部的中高层访谈问卷调查,运用探索性因子分析模型,对不良贷款主成因进行萃取和诠释,最后构建银行业不良贷款的成因层次构面,以支持国有商业银行授信决策之依据。     

On the economic order quantity under conditions of permissible delay in payments

In this note, I amend Goyal's model by considering the difference between unit price and unit cost. I then establish an easy analytical closed-form solution to the problem. The theoretical results obtained here reveal the following two managerial phenomena. (1) In certain cases, the economic replenishment interval and order quantity decreases under the permissible delay in payments, which contradicts to Goyal's conclusion. It makes economic sense for some customers to order less quantity (or shorten the replenishment time interval) and to take the benefits of the permissible delay more frequently. (2) If a supplier wants to reduce his/her large level of inventory, then he/she should charge an excessive interest rate on his/her customer's outstanding amount after the credit term expires. Consequently, his/her customers will order to buy more quantity than the classical economic order quantity. As a matter of fact, these two managerial phenomena have been demonstrated in the decision making of using credit cards. For example, most credit card companies provide card holders 25 days of grace period, and charge 18–20% interest on the amount past due (ie, the second phenomenon). However, for a well-established credit card holder, he/she will take the benefit of 25 days of grace period constantly, but will not spend over his/her limit and face an excessive finance charge (ie, the first phenomenon).     

Optimal allocation between bank loans and treasuries with regret

The main categories of assets held by banks are loans, Treasuries (bonds issued by the national Treasury), reserves and intangible assets. In our contribution, we investigate the investment of bank funds in loans and Treasuries with the aim of generating an optimal final fund level. Our results take behavioral aspects such as risk and regret into account. More specifically, we apply a branch of optimization theory that enables us to consider a regret attribute alongside a risk component as an integral part of the utility function. In this case, regret-aversion corresponds to the convexity of the regret function and the bank’s preference is assumed to be representable by optimization subject to the utility. In addition, we provide a comparison between risk- and regret-averse banks in terms of optimal asset allocation between loans and Treasuries. A feature of our contribution is that these and other optimization issues are analyzed briefly and, where possible, represented graphically. Furthermore, we comment on the claim that an investment away from loans towards Treasuries is responsible for credit crunches in the banking industry.     

Modelling credit risk of portfolio of consumer loans

One of the issues that the Basel Accord highlighted was that, though techniques for estimating the probability of default and hence the credit risk of loans to individual consumers are well established, there were no models for the credit risk of portfolios of such loans. Motivated by the reduced form models for credit risk in corporate lending, we seek to exploit the obvious parallels between behavioural scores and the ratings ascribed to corporate bonds to build consumer-lending equivalents. We incorporate both consumer-specific ratings and macroeconomic factors in the framework of Cox Proportional Hazard models. Our results show that default intensities of consumers are significantly influenced by macro factors. Such models then can be used as the basis for simulation approaches to estimate the credit risk of portfolios of consumer loans.     

基于模糊子集的贷款组合优化方法

遵照国际银行业大多数银行的做法,信用风险评估包括对债务人和债项两个方面.以模糊集理论为基础,通过试算与比较,构造隶属函数,对各指标进行无量纲化处理,建立距离判别函数,评估债务人信用风险.根据债项特征,考察风险四因素:违约概率,特定违约损失,违约敞口,期限,建立0-1整数规划模型,对债项进行风险评估,确定最佳贷款组合,以解决组合贷款的优化决策问题.     

Customer retention: a behavioural model

One of the main problems currently facing CTedit-card issuersis the increasing number of cardholders who are using theircards less often (i.e. attrition) and/or returning their cards(closures). This problem is of particular concern as the totalnumber of credit cards held by consumers is declining (by approximately0.6 per cent per month in 1992) and the number of new applicantsis also running at an all time-low (less than 1 per cent permonth in 1992). Most of the published literature in the broad area of creditcards looks at credit scoring, rather than the need for cardissuers to identify and retain a profitable portfolio of cardcustomers. The overall objective of our research is ‘segmentationfor customer retention’, and this paper aims to identifythe characteristics of card customers who initiate the closureof their accounts Linear discriminant analysis is applied toa sample of approximately 17,000 UK holders of bank credit cards,using various behavioural and sociodemographic variables, andtested on a holdout sample of 10,000 cases.     

基于综合风险收益的贷款组合优化决策

商业银行货款组合决策的过程,是遵循“效益性、安全性、流动性”的原则,在综合考虑贷款收益和风险的前提下,从众多的贷款对象中选择一组合适的贷款对象的过程。建立综合考虑贷款收益和风险的贷款决策模型,有利于银行通过量化计算进行科学决策,以提高信贷质量,达到商业银行的经营目标。     

A comparison of neural networks and linear scoring models in the credit union environment

The purpose of the present paper is to explore the ability of neural networks such as multilayer perceptrons and modular neural networks, and traditional techniques such as linear discriminant analysis and logistic regression, in building credit scoring models in the credit union environment. Also, since funding and small sample size often preclude the use of customized credit scoring models at small credit unions, we investigate the performance of generic models and compare them with customized models. Our results indicate that customized neural networks offer a very promising avenue if the measure of performance is percentage of bad loans correctly classified. However, if the measure of performance is percentage of good and bad loans correctly classified, logistic regression models are comparable to the neural networks approach. The performance of generic models was not as good as the customized models, particularly when it came to correctly classifying bad loans. Although we found significant differences in the results for the three credit unions, our modular neural network could not accommodate these differences, indicating that more innovative architectures might be necessary for building effective generic models.