Designing a Hybrid Intelligent Mining System for Credit Risk Evaluation

In this study, a novel hybrid intelligent mining system integrating rough sets theory and support vector machines is developed to extract efficiently association rules from original information table for credit risk evaluation and analysis. In the proposed hybrid intelligent system, support vector machines are used as a tool to extract typical features and filter its noise, which are different from the previous studies where rough sets were only used as a preprocessor for support vector machines. Such an approach could reduce the information table and generate the final knowledge from the reduced information table by rough sets. Therefore, the proposed hybrid intelligent system overcomes the difficulty of extracting rules from a trained support vector machine classifier and possesses the robustness which is lacking for rough-set-based approaches. In addition, the effectiveness of the proposed hybrid intelligent system is illustrated with two real-world credit datasets.     

不允许卖空的多因素证券组合投资决策模型

利用套利定价理论(APT)改进不允许卖空的Markowitz的证券组合投资决策模型,导出了不允许卖空的多因素证券组合投资决策模型,并研究了该模型的解及其性质.     

一种综合评价指标属性值规一量化技术新途径

对现有的评价指标属性值规一量化技术进行了论述 ,运用模式识别、人工神经网络聚类理论结合模糊集值统计理论 ,提出了评价指标属性值规一量化技术新途径 ,奠定了评价指标属性值规一量化技术智能化的基础 .     

风险项目投资决策与实物期权估价方法

讨论了实物期权的存在对风险项目投资决策的影响,分析传统的净现值方法与期权方法的联系,提出基于期权定价方法的改进净现值模型,并通过一个扩大投资问题研究了成长期权的价值。     

极值理论(EVT)方法用于受险价值(VaR)计算的实证比较与分析

讨论了根据极值理论 ( EVT)计算受险价值 ( Va R)的两类不同的方法 :基于矩估计的“两次子样试算法”和极大似然估计法 ,并给出了各自理论推导过程和计算步骤 .同时 ,把这两类方法与正态分布和经验分布的结果进行了比较 .应用四种汇率历史数据进行的实证计算表明 ,在极端条件下 ,用极值理论方法估计 Va R具有很高的准确性 ,而矩估计法的结果又优于极大似然估计法 .     

Models of Choosing the Financial Compensation Strategy for Flight Delay

At present, most airlines adopted generally the same amount of compensa- tion strategy when needing to provide financial compensation to all flight delay passengers. However, due to the existence of differences in travel time value, ticket fare, as well as the expectation of compensation for flight delays, the gap between the effect of same amount of compensation and many passengers＇ （especially the high-value ones） expectations is large, it results in that airlines need to pay higher cost of compensation, but the total effect of compensation for passengers are not better. This paper advanced four financial com- pensation strategies for flight delays, summarized their own characteristics, and took into account the interests of both airlines and passengers, built the optimization models of the four financial compensation strategies under the restriction of the airline＇s compensation cost and on the goal of the maximum total effectiveness of the financial compensation to all passengers. Finally, based on the specific circumstances of the flight delays, the paper discussed the method for airline to choose the optimal financial compensation strategy through solving four models and comparing the compensation effectiveness.     

一种基于奇异值分解的非线性滤波新算法

提出一种基于奇异值分解的无导数卡尔曼非线性滤波新算法对UKF算法进行改进.该算法利用奇异值分解作为工具,将原算法中的协方差矩阵进行奇异值分解,可以在一定程度上避免在递推过程中,由于计算误差和舍入误差而引起的协方差矩阵失去正定性,从而导致算法失效的问题.在不降低滤波精度,不增加算法复杂度的前提下,新算法具有很好的数值稳定性.实例仿真结果验证了本方法的有效性.     

仿真逼真度对配送中心作业效率评估的影响

仿真逼真度对配送中心规划方案效率评估有着显著影响.构建了虚拟物理模型、输入数据建模和作业流程建模三维度逼真度层次划分框架,给出仿真代价和仿真效用分析框架,探讨了不同逼真度对仿真代价和仿真效用的影响评价方法,并以某图书配送中心两个作业模块为实例,分别建立了多组三维度不同逼真度组合的仿真模型进行分析评价,最后总结了仿真逼真度对效率评估影响的基本特征和仿真实施前事先确定仿真逼真度的一般方法.     

复杂产品开发过程仿真与预测研究

为了更有效的对复杂产品开发过程进行预测分析,提出一种两层设计结构矩阵模型作为仿真的结构关系模型.建立了服从三角分布的时间与成本模型,采用拉丁超立方抽样法对该模型进行抽样并生成样本点.采用任务返工概率、返工影响与学习效应曲线来描述开发过程的迭代特征,对任务返工概率的变化调整、返工需求判断策略、多任务返工执行约束等问题给出了相应的仿真模拟方法.在此基础上提出了改进的产品开发过程仿真算法,并设计开发了基于Matlab的复杂产品开发过程仿真平台.最后,通过一个实例证明了该仿真平台的有效性和实用性.     

消费品生命周期隐患识别与风险评价的三维矩阵方法

面对近年我国产品质量安全事故频发、消费者人身安全与健康遭受不同程度的威胁以及出口产品不时被召回或退回的问题,给出了一种基于消费品生命周期和事故致因的隐患识别及表征方法,有助于提升人们对消费品危害规律的认知水平并为消费品风险评价奠定基础. 考虑到事故伤害受体的能力和行为会对消费品缺陷或隐患演化及事故伤害后果产生重要影响,在常规风险评价二维矩阵的基础上,加入风险受体的脆弱性要素,提出了消费品风险评价的三维矩阵方法,能够更好地表征消费者能力与行为等主观要素对消费品隐患事故及危害的影响作用. 以家用电器风险评价为例,基于欧盟非食品类消费品快速预警系统（RAPEX）对中国消费品通报信息分析和专家评价打分,形成了消费品评价三维风险矩阵,分析了家用电器的风险规律,为消费品的风险评价提供了新思路、新方法.