基于三维Copula函数的金融指数相关性分析

给出了三维Copula函数模型中未知参数的估计方法及最优三维Copula函数的选择方法,此构造方法对研究多变量之间的相依性提供了新途径.通过对上证指数、深圳成指及创业板指的历史数据进行实证分析,选出了最优三维Copula函数以描述三者之间的相关性,并分析三者之间的尾部相关性.     

基于极值统计和高维动态C藤Copula的股市行业集成风险计算

股市诸多行业风险之间存在着波动相依性,集成计量多维风险对投资决策意义重大。藤Copula是Copula函数高维化拓展的一个方向,其动态化是新的研究前沿。将极值理论的GPD模型和高维动态C藤Copula方法结合起来研究沪深300指数中地产、基建、银行和运输四个行业风险,能够有效描述尾部极值形态,突出关键变量的作用。再运用动态Pair-Copula分解,刻画高维行业风险变量间的动态关系,以仿真出动态集成风险变量VaR序列。VaR计算结果通过了回溯检验和稳定性测试,表明高维动态C藤Copula模型可以作为风险集成计量的一种新的有效方法。     

基于藤Copula的GAMLSS模型与非寿险准备金评估

在假设各个业务线的增量已决赔款服从伽玛分布、逆高斯分布和对数正态分布的基础上,建立了各个业务线增量已决赔款的GAMLSS模型,并将此模型应用于一组具有明显异方差的车险数据,拟合效果优于均值回归模型.另外,在多个业务线的准备金估计中,不同业务线之间的相依性通过藤Copula函数来描述.用D藤Copula描述相依关系的GAMLSS模型对准备金的评估结果既优于独立假设下的GAMLSS模型和链梯法对准备金的评估结果,同时还刻画了不同业务线之间的尾部相依性.     

时变多元NBS Copula模型与美国股指期货市场可视化相依性分析

基于多元NBS(Normal Birnbaum-Saunders)分布构造了一种新的多元偏斜厚尾Copula,即多元NBS Copula,并进一步采用DCC(Dynamic Conditional Correlation)模型构造了时变NBS Copula模型。以美国道琼斯30指数期货、标准普尔500指数期货和纳斯达克100指数期货为例，可视化分析了收益率序列之间的各种相依特征，比较了DCC-NBS Copula模型与其他一些Copula模型在相依结构拟合上的效果差异。实证结果表明：美国三大股指期货之间的相依结构具有正相依性、厚尾相依性、非对称相依性和时变相依性，其中，NAGARCH模型可以较好地描述收益率序列的动态特征，椭圆Copula优于阿基米德Copula,非对称椭圆Copula优于对称椭圆Copula,厚尾椭圆Copula优于正态Copula,时变椭圆Copula优于静态椭圆Copula。综合来看，DCC-NBSCopula模型是所有模型中对相依结构的拟合效果最优的。     

基于“藤”结构的高维动态Copula的构建

高维化和动态化是当前Copula理论研究和应用的两个重要方向.采用图形建模工具中"藤"的层叠结构,以二元动态Copula取代原有二元静态Copula作为"藤"的节点,将高维Copula建模中"藤"的方法与动态Copula相结合,构造了"动态藤Copula".实证表明,高维动态藤Copula较相应的高维静态藤Copula对数据的概率模型的似然率更高.     

混合Copula模型在中国股市的应用

首先给出了描述相依结构的混合Copula模型,然后给出寻求混合Copula模型的EM算法,最后以中国股市的实际数据进行了实证分析,说明混合Copula模型是可以用来描述中国股市的相依结构.     

基于Copula-ECM-GARCH模型的农产品期货套期保值绩效研究

Copula函数具有可以准确刻画变量间的相依结构、精准描述金融时间序列"尖峰厚尾"分布特点的良好统计性质.针对传统计量模型在计算套期保值比率时存在的局限性,利用Copula函数描述变量的尾部相关性,并结合ECM-GARCH模型,对大豆、小麦、玉米三种国内农产品期货进行套期保值研究,分别计算最优的套期保值比率及其绩效,并...     

多元Copula密度估计的小波局部阈值方法

为了量化资产之间相依结构的局部特征,本文将小波阈值规则引入Copula参数估计,提出多元Copula密度的小波局部阈值估计量,发现Copula密度的光滑度指数、维数和采样容量是影响估值精度的重要因素,这一点也得到了以正态Copula为仿真算例的支持。本方法增强了参数Copula建模的局部自适应能力,进而有助于改进资产的市场风险估值与最优化配置。     

基于Copula函数的屏蔽数据并-串联系统可靠性分析

讨论了失效相依屏蔽数据系统的可靠性分析问题.通过引入Copula函数描述部件寿命变量之间的相依关系,建立屏蔽数据并-串联系统可靠性模型,推导出并-串联系统的一些概率结果.在此基础上,基于逐步Ⅰ型混合截尾的系统失效数据,获得了模型参数及可靠性指标的极大似然估计和bootstrap区间估计.最后,运用蒙特卡罗模拟验证了方法的可行性和有效性.     

藤Copula模型与多资产投资组合VaR预测

投资组合风险管理往往涉及多个资产,在传统的二元Copula函数面临"维度诅咒"问题及多元Copula函数刻画多变量联合分布时其精确性和灵活性存在各种局限性的情况下,引入藤Copula刻画多个资产收益的联合分布,基于不同的Pair-Copula类别构建藤Copula,运用蒙特卡罗模拟方法计算多资产投资组合的VaR,通过Kupiec和Christoffersen返回检验方法测试藤Copula模型的VaR预测效果,并与传统方差-协方差风险管理方法做比较。实证分析表明,传统的方差-协方差风险管理方法和基于正态Pair-Copula作为藤Copula构建模块的方法不能通过多资产投资组合的VaR预测返回检验;而基于student-t Copula、Clayton Copula具有尾部分布特征的Copula作为构建模块的藤Copula模型能够有效地用于多资产投资组合VaR预测,从而更好的用于指导实践。     

野生中药材供给与稽查的不完全信息动态博弈分析

针对中药材供给中的信息不对称性,建立了三个中药材供给方与政府稽查部门之间的三个不完全信息动态博弈模型,得到了相应的子博弈精炼贝叶斯均衡解的五个结论.这五个结论合理刻画了中药材供给方与政府稽查部门的博弈行为并揭示了非法采挖野生中药材现象难以杜绝的原因.结果表明,政府对稽查行为的条件激励措施并不能有效杜绝非法采挖野生中药材现象,只有政府对所有的稽查行为实施普遍的强激励机制才可从根本上消除非法采挖野生中药材的现象.     

Identifying future defaulters: A hierarchical Bayesian method

Traditional methods of applying classification models into the area of credit scoring may ignore the effect from censoring. Survival analysis has been introduced with its ability to deal with censored data. The mixture cure model, one important branch of survival models, is also applied in the context of credit scoring, assuming that the study population is a mixture of never-default and will-default customers.     

综合评价试卷质量分析方法

试卷综合评价分析是提高教学质量的重要环节,也是目前高等院校办学水平和教学质量评估的一项重要的基础性工作.传统的试卷评价分析方法只是一般意义上的正态分布检验,层次是粗浅的.本文将从知识点覆盖率、题量、难度系数、区分度、稳定性和目标效果等方面探索对试卷进行综合评价分析的深层次方法.     

方差分析在复杂测量系统能力分析中的应用

测量系统分析是QS9000标准中的核心要素之一,主要用来评估测量系统对工序质量改进的保证程度。传统的测量系统能力分析主要采用极差分析或方差分析研究单台测量仪器组成的二因素交叉型简单测量系统,本文从企业的实际情况出发,采用方差分析法研究包含多台测量仪器的复杂测量系统,基于三因素套析因混合效应实验模型的方差分析和方差估计,提出了这种测量系统能力分析的方法,并给出了应用实例。     

Heuristic approaches to large-scale periodic packing of irregular shapes on a rectangular sheet

The nesting problem is a two-dimensional cutting and packing problem where the small pieces to cut have irregular shapes. A particular case of the nesting problem occurs when congruent copies of one single shape have to fill, as much as possible, a limited sheet. Traditional approaches to the nesting problem have difficulty to tackle with high number of pieces to place. Additionally, if the orientation of the given shape is not a constraint, the general nesting approaches are not particularly successful. This problem arises in practice in several industrial contexts such as footwear, metalware and furniture. A possible approach is the periodic placement of the shapes, in a lattice way. In this paper, we propose three heuristic approaches to solve this particular case of nesting problems. Experimental results are compared with published results in literature and additional results obtained from new instances are also provided.     

小样本下成本动因相关性的检验——以某固井公司运输作业为例

提高成本相关性的关键是提高成本动因的相关性.传统方法在成本动因选择方面依靠的是规范分析方法。样本匮乏是难以采用实证方法检验成本动因相关性的根本原因.本文通过引入部门ABC获取足够样本,采用实证检验方法研究成本动因的选择问题:首先,在实地获取第一手资料的基础上,通过规范分析,确定备选成本动因;其次,引入部门ABC获取足够的样本;最后,采用统计检验方法研究成本动因的相关性选择.研究发现,原有成本动因不是最佳成本动因;ABC方法确定的成本动因是最佳成本动因.     

Farewell to direct source incompatibilism

Traditional theorists about free will and moral responsibility endorse the principle of alternative possibilities (PAP): an agent is morally responsible for an action that she performs only if she can do or could have done otherwise. According to source theorists, PAP is false and an agent is morally responsible for her action only if she is the source of that action. Source incompatibilists accept the source theory but also endorse INC: if determinism is true, then no one is morally responsible for any action. This paper is a critique of a kind of source incompatibilism, namely, direct source incompatibilism. Direct source incompatibilists reject PAP on the basis of Frankfurt-style examples. Since PAP is one of two premises in the traditional argument for INC, direct source incompatibilists opt for a version of the direct argument, which argues for INC with the aid of some non-responsibility transfer principle. I demonstrate that this option is not available, for there is a tension between the following two claims.     

Fashion Forecasting for a Mail Order Company Using a Bayesian Approach

Traditional forecasting methods assume a large amount of product history. New product launches take place in the absence of any product sales statistics and initially it is necessary to formulate beliefs as to the product future and then to combine these with sales data as it becomes available. The particular situation considered in this paper concerns a mail order company which sells a different group of ladies dresses on each of its catalogues. The life of a catalogue is of the order of 6 months and material ordering decisions tend to be relevant only at the beginning and during the first few weeks of the catalogue. The task of the Distribution Department is not helped by the presence of a large number of returns from customers. This paper describes a Bayesian approach to the forecasting problem and the "live" performance of the method on one catalogue is discussed.     

Order splitting in single sourcing with scheduled-release orders

Traditional inventory models assume that a buyer places one order with a supplier in each order cycle. A large number of researchers have studied the benefits of dual sourcing such that an order quantity is split and placed simultaneously with two suppliers. We show that many of the benefits of dual sourcing are due to order splitting rather than using two suppliers. We investigate order splitting with one supplier such that the first part of the order is sent out immediately but the second part of the order is released later (scheduled-release). Through extensive computational results, we show that in many situations where dual sourcing or the use of a cheaper supplier would be cost effective, single sourcing with order splitting using scheduled-release orders is better. The paper provides a quantitative rationale to continue with one supplier. We also summarize the qualitative reasons to prefer single sourcing or multiple sourcing.     

A literature review on the application of evolutionary computing to credit scoring

The last years have seen the development of many credit scoring models for assessing the creditworthiness of loan applicants. Traditional credit scoring methodology has involved the use of statistical and mathematical programming techniques such as discriminant analysis, linear and logistic regression, linear and quadratic programming, or decision trees. However, the importance of credit grant decisions for financial institutions has caused growing interest in using a variety of computational intelligence techniques. This paper concentrates on evolutionary computing, which is viewed as one of the most promising paradigms of computational intelligence. Taking into account the synergistic relationship between the communities of Economics and Computer Science, the aim of this paper is to summarize the most recent developments in the application of evolutionary algorithms to credit scoring by means of a thorough review of scientific articles published during the period 2000–2012.