基于判别分析的R＆D项目中止决策

R＆D项目中止决策是R＆D管理过程中一项非常重要的工作，有效、及时的中止决策对于优化科技资源配置、提高创新效率有着非常重要的意义。本文在对已有的R＆D项目中止决策方法简要回顾基础上，根据R＆D项目中止决策实质，给出了一种基于已有数据判别分析的R＆D项目中止决策方法和步骤，并进行了示例计算。结果表明，该方法较为简便有效，易于操作。     

后金融危机时代股票超额收益率影响因素的探索

运用判别分析法和决策树模型对上证180的成分股是否可以获得超额收益率及其影响因素进行了分析,并将两种方法的分析结果进行了对比。结果表明：两种方法都可以对上市公司是否可以获得超额收益进行分辨,并且可以找出影响因素。但是从两种模型的结果来看,决策树模型要明显优于判别分析法。从我们的分析样本来看,对企业是否可以获得超额收益率的影响因素主要有：市盈率、市净率、息税折旧摊销前利润/营业总收入等。在实际中,通过这两种方法,投资者可以寻找出那些具有超额收益率的股票进行投资,从而使得自身的投资结果优于大盘。     

基于判别分析的内部控制重大缺陷预测研究

公司破产风险、财务舞弊事件通常与内部控制重大缺陷存在密切关系,研究内部控制重大缺陷判别方式具有重要意义和价值。本文以深交所2007—2008年138家内部控制重大缺陷公司与138家非重大缺陷公司为样本,采用逐步判别分析法构建内部控制重大缺陷判别模型,帮助监管方、外部投资者等更好地预测内部控制重大缺陷。研究表明,相比非重大缺陷公司,内部控制重大缺陷公司上市年限较长、规模较小、外部审计师变更更加频繁,盈利能力和成长性较差,但流动性、营运能力指标的检验结果是混合性的。交互验证结果表明,模型预测准确率达到70%以上。     

基于Bayes判别法的脑电图数据分析的研究

目的:本文通过对客观记录的受试者脑电图数据进行Bayes判别分析,判断其能否应用于脑电数据特征提取和分类决策。为脑电图研究的其它分析做基础分析。方法:根据α波的强弱不同将21导电极分为四类,分别对63例正常状态下受试者21导联电极的脑电图数据进行Bayes判别分析,并利用误判率回代估计法检验判别准确率。数据处理和统计分析采用独立设计的脑电图分析工具箱和Bayes判别分析程序。结果:表明对63例正常状态下受试者的脑电图数据进行Bayes判别分析,预测各电极分类准确率75.4%。结论:Bayes判别法预测准确率较高,脑电特征(主要为α波)提取较为准确,能较好的应用于脑电数据特征提取和分类决策中,从而辅助脑电图的检查和定量分析,为脑电图的检验提供有效的分析手段。     

Deferred Taxes and Bond Ratings: A Canadian Case

This paper explores the usefulness of the current Canadian Institute of Chartered Accountants standard on accounting for income taxes in bond rating decisions by credit analysts. Bond rating prediction models using accounting variables generated with alternate treatment of income taxes, have been developed. The analysis indicates that additional information presented by the above standard has not contributed significantly to the bond raters' decision making process.     

基于二维非参数特征分析的SAR图像目标识别

在分析传统Fisher线性鉴别分析局限性的基础上,由图像的行信息和列信息提出了两 种形式的二维非参数特征分析 (2DNFA)的特征提取方法,并应用于SAR图像目标的识别。直 接在SAR图像矩阵上使用非参数特征分析提取特征不仅能充分发挥非参数特征分析的性能而 且保留了图像矩阵的结构信息,大大降低了散度矩阵的维数,减小了运算量。使用美国MSTA R计划录取的数据对算法进行了仿真验证,实验结果显示两种形式的二维非参数特征分析在 较低特征维数下的识别率均可以达到98%以上,表明所提方法的有效性和正确性。     

Green human resource management: Development of a valid measurement scale

The objective of the study was to develop a valid measurement scale for green human resource management (HRM). Even though the common practices of green HRM have been presented in much of the literature, the previous studies focused only on a small number of functions in integrating environmental management with HRM. Additionally, the measurement of green HRM practices still calls for empirical validation. The two‐stage methodology of structural equation modeling in AMOS was employed for data analysis. Exploratory factor analysis revealed seven dimensions of the construct measured by 28 items. Confirmatory factor analysis confirmed the factor structure. The measuring instruments revealed convergent and discriminant validity. Several model fit indices indicated the model fitness. The study provided supplementary evidence on the underlying structure of the construct that can be valuable to researchers and practitioners in this area.     

Factors associated with hunter retrieval of deer hit by arrows and shotgun slugs

Abstract

Hunter retrieval of wounded deer was used as a case study to examine applications and limitations of leisure research for the management of recreational behavior. All hunters receiving permits to hunt in the Shiawassee National Wildlife Refuge during November 1–20, 1983, were sent mail questionnaires. A total of 457 (93 percent) of the archery and 143 (95 percent) of the firearm permittees responded after 3 mailings. Archery deer hunters were estimated to have retrieved 29 (43 percent) of the 68 different deer hit by arrows. Shotgun hunters were estimated to have retrieved 42 (81 percent) of 52 different deer hit. Discriminant analysis showed that the retrieval of deer hit with arrows was most likely when the shooter had tagged a large number of deer during former archery hunts, when the deer was hit in a position other than broadside, and when the range of the shot was short. Deer hit with slugs were more likely to be retrieved when hit by hunters with little experience in tagging deer during former firearm hunts, when the shooter also hunted deer with a bow and arrow, and when the shot was taken at deer that were not running. These data were interpreted to suggest how retrieval rates might be increased through regulation, education, and management of the recreational environment. The role of leisure research and theory in providing techniques and a rationale for modifying the behavior of recreationists is discussed.     

National parks in the third world and associated national characteristics

Abstract

For 102 third world countries, presence and absence of national parks is used as the dependent or classification variable in a discriminant analysis. Eleven variables suggestive of diverse tropical savanna habitats contribute to a single canonical discriminant function. Socioeconomic variables are poor contributors to the function. Seventy‐nine of the countries are properly classified in their group by the canonical discriminant function. Misclassification focuses on sub‐Saharan Africa where many countries have established parks under “improper”; conditions, and strong positive association with the function calls attention to Latin America.     

Using a Validation Process to Develop Market Segmentation Based on Travel Motivation for Major Metropolitan Areas

Studies on market segmentation based on travelers' push motivation have been conducted without implementing a statistically validated procedure for establishing the segmented markets. The purpose of this study is to identify validated traveler market segments by examining what travelers' push motivation dimensions are among the heterogeneity of travelers to a major metropolitan area. The six push motivation dimensions were extracted from the 20 motivation items by conducting factor analysis (i.e., Social/Interaction, Excitement/Fun, Relaxation, Sightseeing, Family/Friends, and Sports). Cluster analysis identified the three homogeneous groups of travelers. Subsequent discriminant analysis verified the existence of the three travelers' markets in the metropolitan area. Managerial implications are suggested for the three push motivation‐based segments.