额外资源按需分配方法研究

文献Wei Quanling等^[1]的最后一段提出了这样一类额外资源分配问题：假设有一些额外的投入资源将要分配给全部或者部分决策单元,如果我们希望分配结果对于整个系统（该系统由全部决策单元组成）来说是最有益的,那么应当如何分配这些额外的投入资源?他们同时还指出系统中哪些决策单元能够获得额外的投入资源,这不仅取决于决策单元的效率,还需要考虑规模报酬。针对此类问题,本文提出一种额外资源按需分配方法：首先,综合决策单元的技术效率与其投影点的规模弹性构建发展曲线,并由决策单元的发展曲线获知其投入变动与产出变动的对应关系;然后,把额外资源分成若干等份,依次分配每一个等份的额外资源,每次分配使系统的总产出增加实现最大,直至分配完全部的额外资源,或者分配某一份额外资源不再带来任何的产出增加,分配就会停止;最后,把决策单元在每一次分配过程中获得的额外资源进行累加,得到这个决策单元最终的分配结果。全文的最后分别在"一个投入一个产出"和"两个投入一个产出"情况下给出两个具体的分配算例。分配结果表明,这种按需分配方法不会缩减任意一个决策单元当前占有的投入资源数量,并且还可以有效地在辨识出某种额外资源在规模上和结构上存在的冗余,从而有效地避免不必要的分配。     

A multicriteria sorting approach based on data envelopment analysis for R&D project selection problem

In this paper, multiple criteria sorting methods based on data envelopment analysis (DEA) are developed to evaluate research and development (R&D) projects. The weight intervals of the criteria are obtained from Interval Analytic Hierarchy Process and employed as the assurance region constraints of models. Based on data envelopment analysis, two threshold estimation models, and five assignment models are developed for sorting. In addition to sorting, these models also provide ranking of the projects. The developed approach and the well-known sorting method UTADIS are applied to a real case study to analyze the R&D projects proposed to a grant program executed by a government funding agency in 2009. A five level R&D project selection criteria hierarchy and an assisting point allocation guide are defined to measure and quantify the performance of the projects. In the case study, the developed methods are observed to be more stable than UTADIS.     

Data-driven operations management: organisational implications of the digital transformation in industrial practice

Abstract

The ongoing digital transformation on industry has so far mostly been studied from the perspective of cyber-physical systems solutions as drivers of change. In this paper, we turn the focus to the changes in data management resulting from the introduction of new digital technologies in industry. So far, data processing activities in operations management have usually been organised according to the existing business structures inside and in-between companies. With increasing importance of Big Data in the context of the digital transformation, the opposite will be the case: business structures will evolve based on the potential to develop value streams offered on the basis of new data processing solutions. Based on a review of the extant literature, we identify the general different fields of action for operations management related to data processing. In particular, we explore the impact of Big Data on industrial operations and its organisational implications.     

非线性面板数据聚类方法研究

对于一类变量非线性相关的面板数据,现有的基于线性算法的面板数据聚类方法并不能准确地度量样本间的相似性,且聚类结果的可解释性低。综合考虑变量非线性相关问题及聚类结果可解释性问题,提出一种非线性面板数据的聚类方法,通过非线性核主成分算法实现对样本相似性的测度,并基于混合高斯模型进行样本概率聚类,实证表明该方法的有效性及其对聚类结果的可解释性有所提高。     

Integrated methodology for multiple systems estimation and record linkage using a missing data formulation

There are now three essentially separate literatures on the topics of multiple systems estimation, record linkage, and missing data. But in practice the three are intimately intertwined. For example, record linkage involving multiple data sources for human populations is often carried out with the expressed goal of developing a merged database for multiple system estimation (MSE). Similarly, one way to view both the record linkage and MSE problems is as ones involving the estimation of missing data. This presentation highlights the technical nature of these interrelationships and provides a preliminary effort at their integration.     

Copula-based regression models: A survey

In this review paper we collect several results about copula-based models, especially concerning regression models, by focusing on some insurance applications.     

制造业效率区域差异及其影响因素分析——基于中国省级面板数据的研究

制造业的发展关系到第二产业乃至整个国民经济的发展,制造业的增长是中国工业经济增长的主导力量。本文运用数据包络分析法,测度了制造业各行业及中国各省制造业效率情况,然后利用中国省级制造业效率指标以及专利申请、地区经济发展水平等制造业效率影响因素的外延数据构成面板数据（Panel Data）建立模型。通过实证分析得出：地区经济的发展有利于制造业效率的提升。专利申请与制造业效率呈负相关。     

会计公允值制度对股权结构与绩效关系的影响效应研究——基于中国上市商业银行面板删失异质性数据的分析

 本文基于商业银行的特殊性,考虑会计公允值约束,以招商、深发展、浦发展、华夏、民生、兴业银行等上市股份制商业银行2000-2007年的样本数据,运用删失（Censoring）面板数据建模技术（Panel Data Modelling）,选择适当的控制变量,详尽分析了股权集中度和股东性质2个维度8项指标所描述的商业银行股权结构对商业银行银行3个维度7项结构指标的正面绩效和3项结构指标的负面绩效的影响。删失面板建模实证分析结果表明,在各家银行绩效异质性和会计公允值制度的约束下,本文从绩效构成结构方面,更为深入系统地揭示了股权集中度和股东性质对上市商业银行流动性、盈利性、成长性等正面绩效以及竞争结构、监管压力和资产质量等负面绩效的依存关系;各家上市商业银行绩效在结构方面存在异质性;测度股权结构对商业银行绩效的影响效应时,考虑公允值制度与否,所得结论在结构上有着显著的差异性甚至相反的结论。     

月度工业生产发展速度指标解读：差异比较与经验启示 ——以OECD国家为例

 正确解读统计指标是准确理解和把握统计数据意义的关键所在。本文从指标的名称、定义、计算方法、空间规定和基础指标获得方式等方面,对OECD国家的工业生产发展速度指标进行了全面解读,以期从中得出经验与启示。     

A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations

Correlated data are commonly analyzed using models constructed using population-averaged generalized estimating equations (GEEs). The specification of a population-averaged GEE model includes selection of a structure describing the correlation of repeated measures. Accurate specification of this structure can improve efficiency, whereas the finite-sample estimation of nuisance correlation parameters can inflate the variances of regression parameter estimates. Therefore, correlation structure selection criteria should penalize, or account for, correlation parameter estimation. In this article, we compare recently proposed penalties in terms of their impacts on correlation structure selection and regression parameter estimation, and give practical considerations for data analysts. Supplementary materials for this article are available online.