Approximation of density functions by orthogonal series with grouped data

Summary Letg(x) andf(x) be continuous density function on (a, b) and let {ϕ_j} be a complete orthonormal sequence of functions onL ₂(g), which is the set of squared integrable functions weighted byg on (a, b). Suppose that over (a, b). Given a grouped sample of sizen fromf(x), the paper investigates the asymptotic properties of the restricted maximum likelihood estimator of density, obtained by setting all but the firstm of the ϑ_j’s equal to0. Practical suggestions are given for performing estimation via the use of Fourier and Legendre polynomial series. Research partially supported by: CNR grant, n. 93. 00837. CT10.     

Any complete preference structure without circuit admits an interval representation

This article proves that all complete preference structures where the strict preference relation (P) has no circuit admit a representation by intervals of the real line; the rule for deciding whether an interval is indifferent or preferred to another is less straightforward than for interval orders: strict preference is indeed compatible with a certain degree of overlapping of intervals, the allowed degree being specified by means of a so-called tolerance function.     

Sampling from spatial databases

This paper deals with techniques for obtaining random point samples from spatial databases. We seek random points from a continuous domain (usually ²) which satisfy a spatial predicate that is represented in the database as a collection of polygons. Several applications of spatial sampling (e.g. environmental monitoring, agronomy, forestry, etc) are described. Sampling problems are characterized in terms of two key parameters: coverage (selectivity), and expected stabbing number (overlap). We discuss two fundamental approaches to sampling with spatial predicates, depending on whether we sample first or evaluate the predicate first. The approaches are described in the context of both quadtrees and R-trees, detailing the sample first, acceptance/rejection tree, and partial area tree algorithms. A sequential algorithm, the one-pass spatial reservoir algorithm is also described. The relative performance of the various sampling algorithms is compared and choice of preferred algorithms is suggested. We conclude with a short discussion of possible extensions.     

De minimis and equity in risk

Indices and orderings are developed for evaluating alternative strategies in the management of risk. They reflect the goals of reducing individual and collective risks, of increasing equity, and of assigning priority to the reduction and to the equity of high risks. Individual risk is defined as the (random or non-random) level of exposure to a danger. In particular the role of a lower negligibility level is investigated. A class of indices is proposed which involves two parameters, a negligibility level and a parameter of inequality aversion, and several interpretations of the indices are discussed. We provide a set of eight axioms which are necessary and sufficient for this class of indices, and we present an approach to deal with partial information on the parameters.     

A fast splitting procedure for classification trees

This paper provides a faster method to find the best split at each node when using the CART methodology. The predictability index is proposed as a splitting rule for growing the same classification tree as CART does when using the Gini index of heterogeneity as an impurity measure. A theorem is introduced to show a new property of the index : the for a given predictor has a value not lower than the for any split generated by the predictor. This property is used to make a substantial saving in the time required to generate a classification tree. Three simulation studies are presented in order to show the computational gain in terms of both the number of splits analysed at each node and the CPU time. The proposed splitting algorithm can prove computational efficiency in real data sets as shown in an example.     

品牌资产及其测量中的概念解析

品牌资产(品牌权益)是近年来营销领域中研究的热点问题,然而对于其基本概念的曲解却阻碍了对这个重要问题的进一步研究。本文针对品牌资产涉及的几个概念(品牌资产、品牌权益、品牌价值)及其相互关系加以解释,并将主要的品牌资产的测量方法进行分类,目的是澄清人们对于品牌资产基本概念的模糊认识,为进一步深入研究指明方向。     

Early experiences of mentoring: Design and use of multimedia materials for teaching OR/MS

The MENTOR project, based at the University of Strathclyde, is producing multimedia computer based learning materials with the aim of improving both the effectiveness and the efficiency of teaching Operational Research/Management Science in higher education. The process of development has been carefully designed to encourage ownership of and commitment to the materials amongst lecturing staff in Universities, with the aim of ensuring widespread use. The paper outlines this process of development and design of materials. Experiences in introducing students to the use of these materials as an integrated part of a Management Science class are reported. Feedback from students has been very positive but also indicates the need for ongoing support and direction.     

面向CRM的动态质量体系管理数据的模型

以ISO9000质量管理体系为背景,本文详细地分析了企业质量体系管理的信息层次结构与流程,针对质量体系的制定、改版以及质量持续改进,研究了质量体系管理对数据模型的复用性、可引用性与可回朔性的要求,讨论了现有质量管理信息系统的局限性,提出了动态质量体系管理数据模型,给出了模型的结构和功能,总结了所提模型的特点。最后,给出了实例分析。     

中国上市公司组合治理机制实证研究

一般地,按公司治理机制发挥作用的机理不同,可将其划分为基于公司治理结构的内部治理机制、基于市场的市场治理机制和基于社会环境的社会治理机制。公司绩效是这些治理机制共同作用的结果。本文首先选择设计了15个公司内部与外部治理机制的实证分析指标和1个公司绩效指标:接着以公司绩效为导向,运用数据挖掘技术对截止到2003年3月26日已发布2002年度报的305家沪市上市公司进行实证分析,得到了不同公司治理机制组合与公司绩效的对应关系,这为优化和设置公司治理机制组合、提高公司绩效提供借鉴。     

基于SCD模型的金融超高频数据信息含量研究

针对随机波动条件持续期模型(SCD模型)对于金融超高频数据的模拟优势,采用股票交易的分笔数据,在模型中引入消息指标变量、交易量和买卖价差三类微观结构变量,通过分析价格持续期与这些结构变量之间的相关性,以研究价格形成过程中的信息传递过程。研究结果表明:利好消息通常导致交易强度增加,利空消息通常会导致交易持续期延长,价格变化对交易持续期造成的影响有限;小盘股买卖价差影响后续价格交易持续期,但大盘股的影响并不明显;交易量的增加具有信息含量。