Optimal number of accrual groups and accrual group sizes in longitudinal trials with discrete‐time survival endpoints

In longitudinal trials, the number of accrual groups and their sizes should carefully be chosen to ensure a desired power to detect a specified treatment effect. Methods are proposed to obtain a cost‐effective combination of the number and size of accrual groups that provides high efficiency at minimal cost. We focus on trials where an event occurs at any point in time, but it is recorded on a discrete scale. The Weibull survival function is considered for modeling the underlying time to event. By using a cost function, it is shown that the ratio of the cost of recruiting and treating subjects to the cost of measuring them and also the survival pattern highly influence the optimal combination of the number and size of accrual groups. A maximin approach is further presented to obtain robust designs with respect to poor specification of these modeling parameters. We also show the application of the proposed optimal design methodology using real examples.     

Refined analysis and a hierarchical multi-task learning approach for loan fraud detection

Fraud problems in loan application assessment cause significant losses for finance companies worldwide, and much research has focused on machine learning methods to improve the efficacy of fraud detection in some financial domains. However, diverse information falsification in individual fraud remains one of the most challenging problems in loan applications. To this end, we conducted an empirical study to explore the relationships between various fraud types and analyzed the factors influencing information fabrication. Weak relationships exist among different falsification types, and some essential factors play the same roles in different fraud types. In contrast, others have various or opposing effects on these types of frauds. Based on this finding, we propose a novel hierarchical multi-task learning approach to refine fraud-detection systems. Specifically, we first developed a hierarchical fraud category method to break down this problem into several subtasks according to the information types falsified by customers, reducing fraud identification's difficulty. Second, a heterogeneous network with a meta-path-based random walk and heterogeneous skip-gram model can solve the representation learning problem owing to the sophisticated relationships among the applicants' information. Furthermore, the final subtasks can be predicted using a multi-task learning approach with two prediction layers. The first layer provides the probabilities of general fraud categories as auxiliary information for the second layer, which is for specific subtask prediction. Finally, we conducted extensive experiments based on a real-world dataset to demonstrate the effectiveness of the proposed approach.     

简述PLC可编程控制器在北京南站能源站BAS系统中的选型

文章通过对北京南站能源站"热、电、冷三联供+污水源热泵+补燃"系统BAS专业二次深化设计的总结,从输入输出(I/O)点数的估算,存储器容量的估算、控制功能的选择,机型的选择等方面,介绍BAS系统PLC应用的要点.     

Robust and stable multi-task manufacturing scheduling with uncertainties using a two-stage extended genetic algorithm

ABSTRACT

In cloud manufacturing (CMfg), unexpected uncertainties can occur in real-world manufacturing processes that could make the predetermined schedule infeasible. A new multi-objective proactive method is proposed in this situation to evaluate the proactive schedule. A novel two-stage extended genetic algorithm (2S-EGA) is proposed to generate proactive schedules that consider service interruptions. The experimental results confirmed that the obtained proactive schedule produces great performance when applied to multi-task scheduling problems with service interruptions. Furthermore, the results also showed that the proactive schedule obtained by the proposed approach is more robust and stable than other baseline algorithms taken from the literature.     

简述PLC可编程控制器在北京南站能源站BAS系统中的选型

文章通过对北京南站能源站"热、电、冷三联供+污水源热泵+补燃"系统BAS专业二次深化设计的总结,从输入输出(I/O)点数的估算、存储器容量的估算、控制功能的选择、机型的选择等方面,介绍BAS系统PLC应用的要点。     

Studentization and prediction in a multivariate normal setting

In a simple multivariate normal prediction setting, derivation of a predictive distribution can flow from formal Bayes arguments as well as pivoting arguments. We look at two special cases and show that the classical invariant predictive distribution is based on a pivot whose sampling distribution depends on the parameter – that is, the pivot is not an ancillary statistic. In contrast, a predictive distribution derived by a structural argument is based on a pivot with a parameter free distribution (an ancillary statistic). The classical procedure is formal Bayes for the Jeffreys prior. Our results show that this procedure does not have a structural or fiducial interpretation.     

Output scaling in a cost function setting

A significant result in production theory is that even under a very weak condition on the technology, the cost function is well defined and linearly homogeneous in factor prices. We could term this homogeneity a scaling law, for an equiproportionate scaling of all factor prices results in an identical scaling of the value of the cost function. The purpose of this paper is to determine sets of conditions under which an analogous result may hold for the output variable in the cost function. Our analysis initially focuses on the cost function of a multioutput technology so that the output variable in the cost function is vector-valued. The results we obtain are that some output-scaling laws exist if and only if the underlying technology is ray-homothetic. In the final section we consider the meaning of the results for the single-output case.The refereeing process of this paper was handled through R. Robert Russell. An earlier version of this paper was presented at the 1989 Conference on Current Issues in Productivity, December 6–8, 1989, Rutgers University, Newark, NJ.     

Accountability in an agency model: Project selection,effort incentives,and contract design*

We analyze an agency model of project choice and implementation where the agent is held accountable for his performance. We show that implementation of the ex ante efficient project may be impossible, irrespective of how the principal sets fixed wage and bonus rate. If it is possible, the principal may be forced to increase the bonus rate above the optimal project‐specific rate. The higher profit share compensates the agent for pressure he faces when he has to justify/explain his performance.     

建设项目全寿命周期内工程造价的投资控制

工程造价偏高是造成我国高房价的重要原因,本文指出在建筑全寿命周期内影响工程造价的若干因素,分析其原因,得出我国建筑业的建造模式是制约和影响工程造价的主要原因,通过结合我国建筑企业在国际市场上的竞争力,指出改变传统建筑业的建造模式,将建筑物的工程造价控制在合理的范围内。     

Relating focus to quality and cost in a healthcare setting

Focus in healthcare has been heralded as the next frontier in improving its efficiency and efficacy (Herzlinger 2004). Focus takes several different forms, ranging from standalone specialty centers to a hospital that places a strategic emphasis on a clinical area. We adopt this latter perspective and define focus as a disproportionate emphasis on a particular clinical area in a hospital. We use secondary data from hospitals providing cardiology care in New York State to examine the relationship between focus and performance. We develop two measures of focus. Proportional focus is defined to be the proportion of cases treated in a particular clinical specialty. Expertise focus is defined to be specific evidence that a hospital has taken action to build expertise in treating diseases in that specialty. We operationalize hospital performance along cost and quality dimensions, and we use hierarchical regression to examine the impact of focus on performance. Our results indicate that proportional focus, but not expertise focus, is associated with better cost performance. Quality performance, on the other hand, was associated only with the interaction between proportional focus and expertise focus, which means that only hospitals exhibiting higher levels of both proportional and expertise focus achieve better quality performance. These findings support the notion that not only is focus important in healthcare, but also that researchers and practitioners need to recognize that relationships are contingent on the performance and focus measures used and thus, findings may not be generalizable from one metric to another.     

Reviewing Oates' insights in a repeated game setting

This paper analyzes the optimal assignment of public good policies to layers of a federal system in a repeated game setting. Under a centralized regime, public goods are financed jointly across regions, and a federal legislature decides on the regional quantities. Under a decentralized regime, public goods are financed locally, and governments play a non-cooperative provision game. We find that a centralized (decentralized) regime is more likely to provide the efficient public good policies in case spillovers are small (large). Received: September 2003, Accepted: October 2004 JEL Classification: H11, H41 I wish to thank Clemens Fuest, Anke Kessler, Christoph Lülfesmann, and an anonymous referee for valuable comments on this paper. Financial support by the DFG (SPP 1142) is gratefully acknowledged.     

Quasi-cournot behavior in a multimarket,multiplant setting

This paper introduces multimarket selling options, capacity constraints and quasi-Cournot conjectures into oligopolistic rivalry. It is shown that increased output responses between rivals that are producing below capacity will effect output reductions for all markets, but that increased output responses between rivals that are producing at capacity will effect an output increase in at least one market and might even result in a capacity extension and output increases in all markets.     

System redesign: planning in a multihospital setting

This article explores the complexities of the proposed health care reform and the methodologies by which hospitals are coping with the impending changes. The article focuses on a large multihospital system in Alabama and the reorganization in process at that facility. Centralization as a key to more efficient management, as well as the partnering with smaller hospitals, is highlighted as an approach to dealing with health care reform.     

Portfolio selection in a data-rich environment

We model portfolio weights as a function of latent factors that summarize the information in a large number of economic variables. This approach (hereafter diffusion index approach) offers the opportunity to exploit a much richer information base to improve portfolio selection. We use factor analysis to estimate the space spanned by the factors. This provides consistent estimates for the optimal weights as the number of economic variables and sample size go to infinity. We consider an empirical application to illustrate the practical usefulness of our approach. The results indicate that the diffusion index approach helps to improve the portfolio performance.     

如何转换被审计单位电子数据

审计人员在开展计算机审计工作中,首要问题是对被审计单位的电子数据进行转换,做好电子数据的转换工作是开展计算机审计的前提条件,也是实施计算机审计的关键,我们在对某商业银行审计中,灵活运用《通用审计系统》软件,对其大型会计电算化系统电子数据的转换进行了大胆的探索和尝试,现简要介绍利用《通用审计系统》软件完成电子数据的下载、转换的基本方法。     

Personal and regional redistribution through public finance in a federal setting

Following a partial equilibrium approach, this paper studies the effect of fiscal policy on income distribution in Argentina, based on budget information corresponding to the year 2004. Specifically, it aims to provide an empirical answer to a set of questions related to the responsibility of different levels of government (national or provincial) in ensuring the equitable distribution of income; how this responsibility is accomplished in practice, when this function is shared among two or more levels of government; whether an incompatibility arises among the policies from different levels of government; and the effect of fiscal policy on personal and regional income distribution.The main results of the paper are summarized as follows. At the aggregate level, both national and provincial budgets have a redistributive impact on personal income distribution, through a combination of progressive expenditures and (slightly) regressive taxes. Regional redistribution depends on two fiscal tools: the national budget and the revenue sharing regime. The progressive effect of expenditures and taxes interacts with the geographical effect of the revenue-sharing and the national budget, reinforcing progressivity in net-receiving groups and creating a trade-off between progressivity and (negative) regional transfer in net-financing ones. Provincial budgets have more impact than the national budget, both being compatible in pursuing the redistributive goal.     

Imitation by price and quantity setting firms in a differentiated market

We study the evolution of imitation behaviour in a differentiated market where firms are located equidistantly on a (Salop) circle. Firms choose price and quantity simultaneously, leaving open the possibility for non-market-clearing outcomes. The strategy of the most successful firm is imitated. Behaviour in the stochastically stable outcome depends on the level of market differentiation and corresponds exactly with the Nash equilibrium outcome of the underlying stage game. For high level of differentiation, firms end up at the monopoly outcome. For intermediate level of differentiation, they gravitate to a “mutually non-aggressive” outcome where price is higher than the monopoly price. For low level of differentiation, firms price at a mark-up above the marginal cost. Market-clearing always results endogenously.     

Estimation of volatility in a high-frequency setting: a short review

Decisions in Economics and Finance - Our aim is to give an overview of the topic of estimation of volatility, in a high-frequency setting. We emphasize the various possible situations, relative to...     

Relational union organising in a healthcare setting: a qualitative study

This article explores the impact of a relational organising model within a local UNISON NHS branch. While initial outcomes were modest, we argue that relational approaches have the potential to increase branch engagement with organising activity and to provide a focus on the importance of community within the workplace.