商业银行的个人信用评估研究:基于PCA-GA-BP算法

随着个人消费贷款的普及,贷款人的个人信用评估变得尤为重要。本文选取德国和澳大利亚某商业银行的个人信贷数据为样本数据,采用主成分分析提取样本数据的主成分,通过遗传算法优化神经网络的网络结构、初始连接权值和阀值,然后将优化的神经网络算法用于个人信用评估。与其他算法的准确率比较的结果表明,基于主成分分析—遗传算法—神经网络算法的个人信用评估准确率要高,而且模型的网络结构得到优化,运算时间也有缩短。     

龙海市县域农用地整理规划

研究目的:基于农用地质量以及自组织映射神经网络空间聚类模型,研究县域农用地整理规划。研究方法:主成分分析法,自组织映射神经网络空间聚类模型。研究结果:(1)规划近期农用地整理重点应放在农用地质量等级较好的Ⅱ级与Ⅲ级区域,主要分布于角美、海澄、东园等乡镇;(2)规划中期则以农用地质量Ⅳ级区域为主要整理区,分布于白水、东泗、浮宫等乡镇;(3)规划远期以质量等级差的Ⅴ级农用地为规划区域,分布于港尾、隆教、程溪等乡镇的丘陵山地区。研究结论:基于农用地质量PCA主因子和自组织映射神经网络空间聚类模型的农用地规划能客观反映农用地质量的区域差异性,对县域农用地整理规划、精细型基本农田整理规划以及整理重点项目立项等工作均有借鉴意义。     

基于主成分分析法的员工工作满意度分析

针对A企业的员工工作满意度进行了问卷调查,并运用统计学的主成分分析法和层次分析法对问卷回收的相关数据进行分析。结果表明:①A企业员工对其个人发展前景感到满意的比重高于对企业发展前景、工作环境以及收入分配感到满意的比重;②A企业员工对工作感到基本满意的比重只有一半左右。这说明A企业在创造更好的工作条件来提高员工的满意度方面还有很长的路要走。     

我国宏观经济周期性特征研究——基于高维机制转换因子模型的考察

本文利用高维机制转换因子模型（LD RS FM）研究大规模经济数据的机制转换特征和经济的周期性特征。借助主成分分析（PCA）和共同因子自回归的二步分析方法,LD RS FM从大规模变量中提炼出维数较小并可以概括经济周期运动的共同因子,在此基础上进行机制转换分析。这些共同因子代表了大部分宏观经济运动的趋势和特征,并具有明显的结构化含义。实证结果表明,LD RS FM在中国宏观经济周期性特征研究方面具有一定的理论和应用价值。     

Impact of adoption of a decision algorithm including PCA3 for repeat biopsy on the costs for prostate cancer diagnosis in France

Abstract

Objective:

A recent expert study (RAND Appropriateness Method (RAM)) including a panel of 12 European urologists reported that the PCA3 score may be instrumental in taking appropriate prostate biopsy (PBx) decisions, mainly for repeat PBx. This study determined the cost/benefit balance of introducing PCA3 in the decision-making for PBx in France.

Methods:

Two RAM models, without and with PCA3, were retrospectively applied to a sample of 808 French men who had PBx in 2010 (78% first, 22% repeat). Outcome measures included the proportion of PBx that could have been avoided (i.e., judged inappropriate) in the French sample according to both RAM models, and the estimated impact of application of these models on the annual number of PBx and associated costs for France (based on most recent published data).

Results:

Complete profiles were available for 698 men. In the model without PCA3, 2% of PBx were deemed inappropriate. Knowledge of PCA3 would have avoided another 7% of PBx. Repeat PBx would have been avoided in 5% of cases without PCA3 and in 37% with PCA3. For France, application of the RAM model including PCA3 would result in 18,345 fewer repeat PBx. It would be budget-neutral in the unlikely hypothesis of no complications or no costs incurred by complications and would save €1.7 million for a mean cost for complications of €100/procedure or €5 million for a mean cost for complications of €280/procedure, calculated based on US and Canadian data.

Limitations:

Limitations of the study are the theoretical nature of the analysis and the fact that PCA3 distributions had to be derived from other sources.

Conclusions:

Adoption of RAM expert recommendations including PCA3 for repeat PBx decisions in clinical practice in France would reduce the number of repeat PBx and control costs.     

中国海外上市公司的PCA—SVM财务危机预警研究

在支持向量机（Support Vector Machine,SVM）方法的基础上融入主成分分析（Principal Component Analysis,PCA）方法,可构建PCA—SVM财务危机预警模型。以我国海外上市公司为研究对象,运用PCA提取出对财务危机具有显著影响的特征指标,进而通过训练集在不同核函数下对SVM进行训练,最后运用测试集对经过训练得到的SVM财务危机预警模型进行性能验证与评价。实证研究结果表明,PCA．SVM财务危机预警模型在线性、多项式、径向基和sigmoid四种核函数下都具有良好的预测能力,而径向基核函数下的PCA-SVM财务危机预警模型具有更加优越的学习能力与泛化推广能力。     

结合多源特征与高斯过程模型的SAR图像目标识别

针对合成孔径雷达（Synthetic Aperture Radar,SAR）图像目标识别问题,提出结合多源特征和高斯过程模型的方法。分别利用主成分分析（Principal Component Analysis,PCA）、非负矩阵分解（Non-negative Matrix Factorization,NMF）以及单演信号提取SAR图像的特征矢量,并将它们串接为单一矢量。三类特征从不同角度描述SAR图像目标特性,从而为目标识别提供更为有效的信息。决策分类过程采用高斯过程模型进行多元分类,基于融合特征矢量获得概率意义上的最佳决策。实验中,采用MSTAR数据集设置3类目标、10类目标、型号差异以及俯仰角差异识别问题,结果验证了提出方法的优越性能。     

中国松香区域竞争力研究

采用PNP,PCA,FMR和FR指标对1998～2007年中国松香区域竞争力进行实证分析,结果表明:中国松香产地集中度偏高,主产地松香产量悬殊较大,福建和广东松香产量国内占有率呈下降之势;广西、云南和广东松香产能具有较强比较优势,广西、云南和江西松香比较优势呈增长之势;广西和云南松香深加工程度偏低,福建、江西和湖南松香深加工发展迅猛,广东松香深加工能力呈快速下降之势。     

The kernel trick for nonlinear factor modeling

Factor modeling is a powerful statistical technique that permits common dynamics to be captured in a large panel of data with a few latent variables, or factors, thus alleviating the curse of dimensionality. Despite its popularity and widespread use for various applications ranging from genomics to finance, this methodology has predominantly remained linear. This study estimates factors nonlinearly through the kernel method, which allows for flexible nonlinearities while still avoiding the curse of dimensionality. We focus on factor-augmented forecasting of a single time series in a high-dimensional setting, known as diffusion index forecasting in macroeconomics literature. Our main contribution is twofold. First, we show that the proposed estimator is consistent and it nests the linear principal component analysis estimator as well as some nonlinear estimators introduced in the literature as specific examples. Second, our empirical application to a classical macroeconomic dataset demonstrates that this approach can offer substantial advantages over mainstream methods.     

基于主成分分析法和熵值法结合的供应商评价研究

供应商评价是预防供应中断发生的有效措施,从供应商风险评价的角度讨论了供应中断风险管理中供应商选择的重要性,建立了供应中断风险下供应商评价的指标体系,在此基础上,结合主成分法与熵值法计算供应商的得分并进行排序,最终得到供应商的优劣等级,为供应商的选择提供理论依据.