Bayesian dynamic linear modeling for exploring the impact of recent financial crisis on Japan Credit Default Swap market

This paper investigates the effects of recent subprime financial crisis on Japan Credit Default Swap (CDS) market. We first analyze the relationship between the log return series of the reference rates of a CDS contract and the hazard rate. This provides a theoretical foundation for the use of correlation of the log CDS returns as a representation of credit risk correlation. In the dynamic Bayesian linear modeling framework, we consider an algorithm that allow us to obtain dynamic Bayesian updates for the correlation among the reference rates of an underlying CDS contract. Data from the Japan CDS market is analyzed using the proposed methodology. An empirical analyses on the data segmented by different economic environments are carried out. Results indicate that the estimated implied default correlation captures market structure very well and provides useful information for credit risk management.     

Discrete Time Non-Homogeneous Semi-Markov Reliability Transition Credit Risk Models and the Default Distribution Functions

This paper shows how to apply discrete time non-homogeneous semi-Markov processes (DTNHSMP) with an age index to credit risk. The idea is to consider the credit risk problem as a reliability model indexed by the age and in this way, many semi-Markov results could be adapted to describe credit risk problem. The default state is seen as a “non working state”. As the semi-Markov process is a generalization of Markov process, the presented model can be seen as a more general migration model. In fact, in semi-Markov environment the distribution function (d.f.) of the waiting time before a transition can be of any type and without the strong constraints of the Markov model. Furthermore, some results on the asymptotic behavior of the presented model are given. This permits the construction of the d.f. of the default random variable for each “working state”. An example, constructed manipulating some Standard & Poor’s (S&P) data, is presented.     

随机Petri网性能计算软件关键技术的研究

性能计算对随机Petri网模型的性能分析具有重要作用。以研究随机Petri网性能计算软件关键技术为主要目的,提出了一种独立于计算（CIM）的性能计算软件体系结构元模型,并以此元模型为依据,遵循面向对象的系统分析与设计方法给出该软件独立于平台（PIM）的软件体系结构模型,对核心算法设计方案采用先计算基础性能数据,再计算扩展性能数据的策略分步骤完成,并给出了基于MVC模式实现的PSM模型实例和一个电子农务应用案例。最后,给出了同类性能计算软件的设计可适用此开发方法的初步结论。     

Stochastic Models in the Problems of Predicting the Epidemiological Situation

Cybernetics and Systems Analysis - The paper investigates some stochastic models with discrete and continuous time to solve important problems of predicting the spread of epidemiological diseases...     

基于BP神经网络的信用卡违约风险预测

利用BP(Back Propagation)神经网络模型实现信用卡申请人自然信息对违约风险的预测,确定了违约风险的指标体系,通过对小样本数据的训练和仿真,实现了模型输出与目标输出的高度吻合.     

On the Use of Enumeration for Investigating the Performance of Hypothesis Tests for Economic Models with a Discrete Response Variable

This article notes that it is now practical to use the method of enumerationto analyse the performance of estimators and hypothesis tests of fullyparametric binary data models. The general method is presented and thenemployed to investigate the power performance of a common misspecificationtest for the Probit model. The advantages, disadvantages and limitations ofenumeration compared with standard Monte Carlo simulation are thendiscussed. Finally, an example from experimental economics is used todemonstrate that the methodology can also be used in small empirical studies.     

Stochastic Fragments: A Framework for the Exact Reduction of the Stochastic Semantics of Rule-Based Models

In this paper, we propose an abstract interpretation-based framework for reducing the state space of stochastic semantics for protein-protein interaction networks. Our approach consists in quotienting the state space of networks. Yet interestingly, we do not apply the widely-used strong lumpability criterion which imposes that two equivalent states behave similarly with respect to the quotient, but a weak version of it. More precisely, our framework detects and proves some invariants about the dynamics of the system: indeed the quotient of the state space is such that the probability of being in a given state knowing that this state is in a given equivalence class, is an invariant of the semantics. Then we introduce an individual-based stochastic semantics (where each agent is identified by a unique identifier) for the programs of a rule-based language (namely Kappa) and we use our abstraction framework for deriving a sound population-based semantics and a sound fragments-based semantics, which give the distribution of the traces respectively for the number of instances of molecular species and for the number of instances of partially defined molecular species. These partially defined species are chosen automatically thanks to a dependency analysis which is also described in the paper.     

建三江平原C波段雷达回波强度一致性定标与分析

本文基于相关系数法,以建三江与佳木斯两部新一代天气雷达为例,对两部雷达定标前后天气过程基数据进行解析,通过典型天气过程中两部雷达扫描重合地区回波概率密度、网格化散点图、强度分布图等多种图形图表进行回归统计分析,验证基于数理统计的相关系数法,对依据雷达气象方程的回波强度定标具有良好的指导效果.     

信用逾期预测中不同机器学习模型对比分析

当前金融机构正在努力应对不良资产的增长问题,在信贷领域借贷逾期预测结果的准确性将直接决定不良资产的规模.为了更好预测借贷人的还款能力,通常会引入数据模型方法,但对于数据样本较少的新业务,单纯用这类数据容易导致模型结果过拟合.本文通过实际案例分析,对小样本业务数据进行相似业务数据补充,并采用随机森林、LightGBM、XGBoost、DNN和TrAdaBoost迁移学习方法,旨在为小样本业务在模型建立过程中样本不足的问题提供一种有效的解决方法.研究结果表明,针对数据量少的产品,结合相似金融业务数据后采用这五种机器学习模型方法的预测结果 AUC (area under curve)均大于80,其中使用迁移学习模型比LightGBM、XGBoost、DNN和随机森林模型在预测集上的AUC至少高出2个点;此外迁移学习模型的预测结果的精准率(88%)和召回率(73%)也是最高的.     

MANET中节点的运动模型和性能分析

移动自组网的节点运动模型描述了节点的运动情况，采用合适的运动模型对于分析各种协议性能具有重要的意义。该文结合移动自组网络中路由协议的相关属性，提出了邻居节点比率、链路生存时间和平均链路保持时间3种评价指标，并通过大量的模拟实验分析和比较了随机行走运动模型、随机停留运动模型、高斯－马尔可夫运动模型和无边界运动模型4种运动模型，这对于考虑不同运动模型下路由协议的设计十分重要。     

无线传感器网络随机投递传输协议性能分析

无线传感器网络(wireless sensor networks,简称WSNs)通常是资源有限的,且具有较高的链路差错率.在这样的网络环境中,高可靠性的传输协议会消耗过多的能量并产生较大的传输延迟.而另一方面,许多部署在WSNs 环境中的应用可以容忍一定程度的数据包丢失.因此,在这种特殊的网络背景和应用需求环境中,随机投递传输协议应运而生.这种类型的传输协议在有限的网络资源和传输可靠性之间实现了适当的折衷,符合众多传感器网络应用系统的特殊需要.为了对已有的几种随机投递传输协议进行性能分析与评价,采用带吸收态的有限状态马尔可夫链模型对其进行建模.这种建模方法使得协议性能参数的分析计算过程更加直观、简练、易于理解.分析了不同网络参数条件(如源到目的节点之间的跳距,无线链路位差错率等等)对于协议性能的影响.为了提高随机可靠传输协议的效率,部分协议中引入了逐条应答机制或利用无线链路广播特性.分析结果显示,这些措施对协议性能的影响在不同的网络参数条件下也有所不同.最后,为体现分析模型的作用,基于理论分析结果对部分协议进行了改进,并仿真显示改进的效果.     

贝叶斯网络分类模型研究及其在信用评估中的应用

基于概率估计的贝叶斯及贝叶斯网络分类模型,拥有其它数据挖掘工具所不具备的优势。在分析贝叶斯及贝叶斯网络分类模型基础上,结合最小风险决策准则,提出了一种新的信用评估模型。在实际数据集上采用交叉验证方式进行了测试。实验结果表明基于最小风险决策准则的贝叶斯及贝叶斯网络分类模型可以有效地减少信用评估风险。     

基于流随机Petri网模型的可重入生产线稳定性分析

可重入生产线是作为芯片制造生产线模型被研究的，其特点是具有“可重入性”。该文基于流随机Petri网(Fluid Stochastic Petri Net,FSPN)对可重入生产线进行建模，并得到了一些相关的稳定性结论。     

宽带网络中业务模型的性能评价

文中介绍了异步传输模式中的四种业务模型：恒定比特率业务模型CBR、实时可变比特率业务模型rt—VBR和非实时可变比特率业务模型nrt—VBR、可用比特率业务模型ABR，其中详细分析并讨论了非实时的可变比特率和可用比特率两种业务模型，通过对这两种模型进行性能评价以阐明其基本特点。     

宽带网络中业务模型的性能评价

文中介绍了异步传输模式中的四种业务模型:恒定比特率业务模型CBR、实时可变比特率业务模型rt-VBR和非实时可变比特率业务模型nrt-VBR、可用比特率业务模型ABR,其中详细分析并讨论了非实时的可变比特率和可用比特率两种业务模型,通过对这两种模型进行性能评价以阐明其基本特点.     

Performance of a Hedged Stochastic Portfolio Model in the Presence of Extreme Events

Classical methods for computing the value-at-risk(VaR) do not account for the large price variationsobserved in financial markets. The historical methodis subject to event risk and may miss some fundamentalmarket evolution relevant to VaR; thevariance/covariance method tends to underestimate thedistribution tails and Monte Carlo simulation issubject to model risk. These methods are therebyusually completed with analyses derived fromcatastrophe scenarios.We propose a special case of the extreme-valueapproach for computing the value-at-risk of a stochasticmulticurrency portfolio when alternative hedgingstrategies are considered. This approach is able tocover market conditions ranging from the usual VaRenvironment to financial crises.We implement a multistage portfolio model with anexchange rate dynamic with stochastic volatility. Theparameters are estimated by GARCH-t models. Thesimulations are used to select multicurrencyportfolios whose exchange rate risk is hedged andrebalanced each ten days, accounting for VaR. Wecompare the performances of the two most classicalinstitutional options strategies – protective puts andcovered calls – to that of holding an unhedgedportfolio in presence of extreme events.     

评价回归模型拟合效果的数量化方法

回归模型模拟的效果直接影响其预测结果的精确性,针对残差分析法检验模型似合效果的局限性,采用数量化方法进行检验,不仅综合分析的模拟效果,而且可以直接比较模型间的拟合效果,并为建立拟合效果好的新模型提供了充分的依据。     

Investigating the Dynamic Spillover Effects of Low-Cost Airlines on Airport Airfare Through Spatio-Temporal Regression Models

Low airfare is associated with low passengers’ transportation expense, high regional accessibility, and active related industry market. Low-cost airlines have been found to be effective on airfare reduction. However, most previous studies examined their impacts from the perspectives of certain routes and companies. This paper extends the literature by examining the impact of low-cost airlines from the perspective of airports. A spatio-temporal regression model is used to analyze the relationship between airport airfare and low-cost airlines’ market shares. The spatial consideration, which is also under-investigated in the air literature, creates a chance to analyze the spatial interaction of airfare among airports. Along with the temporal consideration, this analytic framework helps to quantitatively show the impact of low-cost airline over time and space by the dynamic spillover-effect function. A case study of market share improvement of AirTran Airway at Albany International Airport (ALB) is used to illustrate the dynamic spillover effect. Results show that the average airfare of ALB will drop $18.64 right after AirTran takes 10 % market share, continue to drop in the following years, and stabilize at around 10 years. Airfare at other airports will also drop, and the magnitudes decrease as time and distance to ALB increase.     

Xen虚拟机Credit调度算法的实时性能分析

为了降低开销以及增加灵活性,通过虚拟化技术将多个系统运行在一个通用计算平台上已成为复杂实时嵌入式系统的趋势。Xen是近年来应用最广泛的虚拟化技术,对其默认使用的Credit调度算法进行实时性能分析,使得能够直接对运行在Xen上的实时系统进行可调度性测试,并且可以通过形式化的资源界限函数对Credit的实时性进行直观的评估。首先分析了Credit调度算法的基本实现,提出并且证明了一种配置VCPU参数的方法使得Credit的实时性得到提升,在此基础上,通过证明得到了Credit算法的基本性质,并得出其在最坏情况下为VCPU分配的资源函数曲线。     

非高斯背景下的频域广义匹配检测

采用广义匹配滤波器在功率谱未知的非高斯噪声中检测宽带信号。在估计功率谱时,为减小周期图法谱估计产生的误差,采用了平滑周期图法来估计非高斯噪声的功率谱,并在此基础上对信号进行恒虚警概率损失下的广义匹配滤波检测。实验证明,用平滑周期图法估计功率谱并进行的广义匹配检测,可以得到较好的检测效果。其检测性能稍弱于功率谱取真值时的检测,但远远优于传统的匹配滤波器。这一结论用到处理复杂的声纳和雷达信号,在实际应用中有一定的指导意义。