基于遗传算法和分类树的信用分类方法

银行信贷信用评估本质上是个分类问题,已有统计和非统计的各种方法应用于信用评估,其中分类树方法,也称为递归分割法,比较适用于处理定性变量,而作为非统计方法之一的遗传算法则适用于处理连续型定量变量之间的非线性关系,但无法处理定性变量,利用这两种方法特点的互补性,构建了一种分类树和遗传算法相结合的信贷信用评估方法,先用分类树方法按照定性变量分类,然后在每个叶结点上用遗传算法按照定量变量分类.实证分析表明,该方法比单独使用分类树方法或遗传算法的分类准确率高.     

包含年龄结构及迁入、迁出地的人口模型

将带有年龄结构的人口迁移模型由只包含迁入地推广到了迁入和迁出地,分析了解的渐近性质,讨论了迁移对迁入及迁出地年龄结构的影响,提供了研究区域内部迁移的框架,如城乡迁移等。     

The use of patient registers to estimate migration

This article presents information discussed at the recent regional seminars on population statistics. It describes a new data source for estimating internal migration between local and health authorities in England and Wales, discusses the way that the data have been collected and processed, and how limitations in the raw data have been addressed. It then goes on to explain the way that the new internal migration estimates have been used in the calculation of population estimates for local and health authorities. This is a change to the method of estimating migration in the calculation of population estimates.     

Families, groups and clusters of local and health authorities: revised for authorities in 1999

This is the second of two articles describing the revision of The ONS classification of local and health authorities of Great Britain to take account of area changes in the 1990's. It presents local authorities in Great Britain as at April 1999 in terms of Families, Groups and Clusters, based on 37 socio-economic and demographic variables from the 1991 Census. Health authorities are presented in terms of Families and Groups only, based on the same variables. The previous article (in Population Trends 98) presented, for each local and health authority, up to four other authorities to which it is 'most similar'.     

集成学习的多分类器动态融合方法研究

AdaBoost集成学习方法中,分类器一经学习成功,其投票权值就已确定,同一分类器对所有待测样本均有相同的投票权值。对于难于分类样本,具有良好分类性能的少数分类器权值却较低。提出适用于集成学习方法的权重自适应调整多分类器集成算法。根据多分类器行为信息,产生待测样本局部分类精度的有效判定区域,基于有效判定区域选择不同的分类器组合,并调整其相应权重,利用样本集上的统计信息来动态指导分类集成判决。实验结果表明,该算法提高了集成分类性能。     

系统方法论的发展及其内在动因分析

在分析“硬”、“软”两类系统方法论建立特点的基础上,探讨了多系统方法论兴起的背景以及整合类系统方法论对上述两类系统方法论提升。指出这种提升的内在动因有3个:复杂性认知动因、系统思想发展的认知动因和文化认知动因,从而为认识新系统方法论提供新的视角,并为理解目前系统方法论的快速发展原因提供了一种解释。     

Which authorities are alike?

This article describes the first results from the recent revision to The ONS classification of local and health authorities of Great Britain to allow for boundary changes. For each local and health authority of Great Britain as at April 1999 it presents 'most similar' authorities as measured by a selection of 37 socioeconomic and demographic variables from the 1991 Census. These results will assist local and health authorities in identifying other appropriate authorities for comparative studies. A second article, to be published in Population Trends 99, will present the results of the analysis to identify groups of authorities in a hierarchy of new Clusters, Groups and Families for the revised classification.     

基于主成分分析和组合神经网络的短时交通流预测方法

提出一种将主成分分析和组合神经网络相结合的方法来预测短时交通流量.将预测路段历史流量及其相关路段的历史流量进行主成分分析,分析结果作为组合神经网络的输入数据,这样不仅减少了输入变量个数,减小网络规模,而且保留了原始变量的主要信息,消除变量之间的相关性.给出了一个说明该方法有效性的例子,用同一组数据比较该方法与典型BP网络的预测效果,结果表明该方法明显优于BP网络.     

改进的逆解耦自抗扰内模控制在磨矿中的应用

针对磨矿分级系统的多变量、强耦合、大时滞等特性,提出改进的逆解耦自抗扰内模控制方法。利用逆解耦方法实现磨矿分级系统的解耦,对解耦后的子系统采用改进的内模控制和线性自抗扰控制。通过引入内模补偿器和增益对时滞进行补偿,减小了系统对模型的依赖度,通过调节线性自抗扰控制器参数、内模补偿器参数和增益来抑制模型失配、外部干扰及不确定性因素对系统带来的不利影响。仿真结果表明,改进的逆解耦自抗扰内模控制方法有较好的解耦性能、跟踪性能和鲁棒性能,验证了此方法是有效的。     

系统发展对策生成的子系统流位反馈环结构分析法——以银河杜仲经济生态系统现代农业区建设为例

系统反馈环结构分析是生成促进系统良性发展对策的一个有效方法. 针对反馈环构建及其运行规律分析缺乏规范方法的问题，建立系统发展对策生成的子系统流位反馈环结构分析法：首先，采用流率基本入树建模法构建系统整体流图，用树枝向量行列式计算出流图全部反馈环；接着，根据系统目标划分子系统，确定各子系统流位变量，构建子系统的流位反馈环结构图；分析确定子系统各反馈环的关键变量，及影响关键变量的调控变量和调控因果链，揭示政策作用的杠杆点；最后，依据杠杆点和调控因果链、遵循整体对策生成原则，确定系统发展对策. 全文以银河杜仲经济生态系统现代农业区建设为实例，演示子系统流位反馈环结构分析方法的基本步骤，生成4项发展对策，对策实施效果证明了该方法的可行性和有效性.     

江西新型农村合作医疗制度实施效应反馈仿真分析

农村居民"看病难、看病贵"问题突出,迫切需要解决.实施新型农村合作医疗(以下简称新农合)正是缓解、解决此问题的关键.为此创建系统动力学入树反馈仿真作用力五步分析法,并应用于江西新农合制度实施效应分析.通过深入调研,确定新农合五个核心变量;建立流位流率系和由流位控制流率的五棵新农合系统流率基本入树,通过顶点和弧并运算,建立系统仿真实验流图模型;然后,构建强简化新农合系统流率基本入树模型,将枝向量代数方法应用到新农合系统的反馈环计算;接着,基于反馈环集合与反馈环特性提出了发展新农合解决"看病贵、看病难"的4条管理对策;最后,通过对新农合制度管理对策的定量仿真实验,得出新农合在农村医药卫生体制改革中产生了四大作用力.     

基于差分进化邻域自适应的大规模多目标算法

对于大规模决策变量给求解大规模多目标优化问题带来的难以收敛及解集分布不均匀问题,通过分析变量特征将其分类再分别优化是当前较为有效的求解方法,但存在变量分类不够准确、变量处理不够有针对性等不足。对此,提出一种基于差分进化邻域自适应策略的大规模多目标优化算法。首先,通过分析扰动解的支配关系将混合变量分为多样性变量和收敛性变量,使变量分类更为准确。其次,通过对收敛性变量主成分分析降噪,降低计算成本,并设计种群的交替进化策略及差分进化的邻域自适应更新操作以提升种群进化过程中的收敛性。实验结果表明,所提算法在收敛速度和解集的分布均匀性上表现出良好的性能。     

Forecasting Approach for Short-term Traffic Flow based on Principal Component Analysis and Combined Neural Network

A combination approach based on Principal Component Analysis (PCA) and Combined Neural Network (CNN) is presented for short-term traffic flow forecasting. The historical data of the forecasted traffic volume and interrelated volumes have been processed by PCA. The results of PCA form the input data for CNN. It not only reduces the dimension of input variables and the size of CNN, but also reserves the main information of the original variables and eliminates relativity among them. An example for explanation of validity is given. The forecast results show that this approach is better than the typical Error Back-Propagation neural network (BP NN) with the same data.     

ULA中基于LS-ICA的短时高效DOA估计

研究了均匀线阵(uniform linear array, ULA)的波达方向(direction of arrival,DOA)估计问题,提出了一种能充分利用阵列结构上来波数据的信号处理方法进行角度估计。所提方法的关键是采用天线阵列因子的实部来构造新的信号接收模型,进而通过独立分量分析(independent component analysis,ICA)对接收信号进行盲分离,用最小二乘(least square,LS)法求解入射信号的角度,同时给出了天线阵元间距应满足的条件。仿真表明,与多重信号分类(multiple signal classification,MUSIC)方法相比,本文方法能够适应阵元间距的变化,在低信噪比(signal-to-noise ratio,SNR)时估计性能以及角度分辨力均较优,通过一步粗估计就可获得较高精度且具有实时性,均方根误差(root mean square error,RMSE)向克拉美-罗界(Cramer-Rao bound,CRB)的逼近程度也说明了新方法的优势,此外还可以采用细估计来提高估计精度。     

The Role of Japanese Candlestick in DVAR Model

The decomposition-based vector autoregressive model(DVAR) provides a new framework for scrutinizing the efficiency of technical analysis in forecasting stock returns. However, its relationships with other technical indicators still remain unknown. This paper investigates the relationships of DVAR model with the Japanese Candlestick indicators using simulations, theoretical explanations and empirical studies. The main finding of this paper is that both lower and upper shadows in Japanese Candlestick Granger contribute to the DVAR model explanation power, and thus, providing useful information for improving the DVAR forecasts. This finding makes sense as it means that the information contained in the lower and upper shadows should be used when modeling the stock returns with DVAR. Empirical studies performed on China SSEC stock index demonstrate that DVAR model with upper and lower shadows as exogenous variables does have informative and valuable out-of-sample forecasts.     

IDENTIFICATION FOR WIENER SYSTEMS WITH INTERNAL NOISE*

This paper considers identification of Wiener systems for which the internal variables and output are corrupted by noises. When the internal noise is a sequence of independent and identically distributed （lid） Gaussian random variables, by the Weierstrass transformation （WT） the system under consideration turns to be a Wiener system without internal noise. The nonlinear part of the latter is nothing else than the WT of the nonlinear function of the original system, while the linear subsystem is the same for both systems before and after WT. Under reasonable conditions, the recursive identification algorithms are proposed for the transformed Wiener system, and strong consistency for the estimates is established. By using the inverse WT the nonparametric estimates for the nonlinearity of the original system are derived, and they are strongly consistent if the nonlinearity in the original system is a polynomial, Similar results also hold in the case where the internal noise is non-Gaussian. Simulation results are fully consistent with the theoretical analysis.     

医方视域的医患信任违背与修复——基于扎根理论的探索性分析

不同于既往从患方视域,论文从医方视域应用扎根理论探讨医患信任违背的深层次因素.结果发现:患方就医善意、就医正直与就医能力、医学固有特征、制度社会情境这5个主范畴对医患信任违背存在显著影响.患方就医形象是医患信任违背的前置内因变量,医学固有特征是医患信任违背的内部情境变量,制度社会因素是医患信任违背的外部情境变量,它们影响医患信任违背的作用和路径并不一致.在此基础上,探索了上述5个主范畴的构成因子及其对医患信任违背的作用机制模型.研究可以为政府制定有效的管控政策以修复医患信任提供理论思路与实施路径.     

基于改进Fishbone的物流仓库布局优化

针对物流仓库内部布局的优化设计问题, 通过在Fishbone布局方法中结合贯通式货架系统的思想, 给出一种改进Fishbone仓库布局方法. 在方法中, 分三种情况对仓库主通道角度和拣货通道角度进行建模, 用Matlab求解并确定了最佳仓库布局角度, 进而考虑托盘尺寸对总移动距离的影响, 建立了最小化总移动距离的数学模型. 结合实际案例结果分析表明, 该方法在基本保持仓库面积利用率的前提下, 有效地缩短了仓库总拣货距离, 不失为一种良好的仓库布局方法.     

ISAR越距离单元走动校正的近似极坐标算法

在ISAR系统中,通常假设不发生越距离单元走动,并应用距离多普勒(RD)算法获得目标的ISAR像。但是当雷达具有较高的分辨率,目标的尺寸较大,或目标在成像相关时间内转过了较大的角度时,散射点的越距离单元走动难以避免。为校正越距离单元走动,提出了一种近似极坐标算法。该算法将相关时间内的雷达回波数据近似为波数域梯形区域上一系列的采样点,并插值得到波数域矩形区域上的采样点,最后通过傅里叶变换得到目标的ISAR像。在目标均匀转动的假设下,该极坐标算法通过近似,不再要求已知目标的转角信息,并且能够较好地校正散射点的越距离单元走动。最后的仿真结果证实了算法的有效性。     

Hierarchy and classification of program plan elements using interpretive structural modeling: A case study of energy conservation in the Indian cement industry

A program can be analyzed and studied by dividing it into elements and subelements. Graphical presentation of the driver power and dependence of the subelements of the identified elements assists in their classification in categories representing the nature of the role each subelement plays as a variable in the program. This paper presents methodology for the hierarchy building of subelements, graphical presentation of their driver power and dependence, and classification in categories such as autonomous, dependent, linkages, and independent variables. The paper further presents a case study of energy conservation in the Indian cement industry to determine the hierarchy of program plan elements and to classify them in categories. Key variables of the program are also highlighted.