Variance estimation based on blocked 3×2 cross-validation in high-dimensional linear regression

In high-dimensional linear regression, the dimension of variables is always greater than the sample size. In this situation, the traditional variance estimation technique based on ordinary least squares constantly exhibits a high bias even under sparsity assumption. One of the major reasons is the high spurious correlation between unobserved realized noise and several predictors. To alleviate this problem, a refitted cross-validation (RCV) method has been proposed in the literature. However, for a complicated model, the RCV exhibits a lower probability that the selected model includes the true model in case of finite samples. This phenomenon may easily result in a large bias of variance estimation. Thus, a model selection method based on the ranks of the frequency of occurrences in six votes from a blocked 3×2 cross-validation is proposed in this study. The proposed method has a considerably larger probability of including the true model in practice than the RCV method. The variance estimation obtained using the model selected by the proposed method also shows a lower bias and a smaller variance. Furthermore, theoretical analysis proves the asymptotic normality property of the proposed variance estimation.     

The Times They are A’changing: Multigenerational Family Simulations

Multigenerational households are increasingly affecting both the individual and family as well as community organizations and social policies. Social work and other family studies students can profit from educational modalities that use adult learning applications through a systems life-course perspective, the whole family aging over time. Family simulation software—addressing multigenerational families, such as two or more adult generations living together—builds on a previous paper (Marriage & Family Review, Feb. 2015). Social class, among other demographic and environmental variables, is emphasized. Agent-based family social network simulation of multigenerational families can facilitate experiential learning. An automatically generated life events report, based on both factual data and specific family characteristics, can be used as a classroom case study for role playing and assessing.     

基于蚁群算法参数的最优选择问题研究

蚁群算法是受现实蚂蚁群体行为启发而得出的一类仿生算法。通过对蚁群算法中影响算法性能的参数进行分析和研究,并对蚁群算法中参数的最优选择问题进行实验分析,从而给出算法参数的最佳取值范围,以利于算法在实际问题中的应用和推广.     

Influencing behaviour: The mindspace way

The ability to influence behaviour is central to many of the key policy challenges in areas such as health, finance and climate change. The usual route to behaviour change in economics and psychology has been to attempt to ‘change minds’ by influencing the way people think through information and incentives. There is, however, increasing evidence to suggest that ‘changing contexts’ by influencing the environments within which people act (in largely automatic ways) can have important effects on behaviour. We present a mnemonic, MINDSPACE, which gathers up the nine most robust effects that influence our behaviour in mostly automatic (rather than deliberate) ways. This framework is being used by policymakers as an accessible summary of the academic literature. To motivate further research and academic scrutiny, we provide some evidence of the effects in action and highlight some of the significant gaps in our knowledge.     

Producing public goods in networks: Some effects of social comparison and endogenous network change

As actors participate in the production of public goods, processes of social comparison may affect the decisions about the individual contributions as well as the endogenously changing structure of an underlying social network. This leads to the transformation of an ordinary n-player Prisoner's Dilemma of collective action into an n-player coordination game. The paper uses agent-based simulations to trace the formation of public goods while varying network characteristics, such as density, segregation, or the strength of relationships. Additionally, the usage of both a forward- and a backward-looking agent model shows possible implications of different assumptions about the actors’ decision making.     

Sex, self-interest and health care priorities

In this questionnaire study, individuals were asked to prioritise publicly provided preventive health care services, one of which would be unavailable to them by virtue of their sex. The aim was to establish whether men and women would exhibit different degrees of self-interest when making a constrained choice. Around 1800 subjects from east-central England prioritised three different types of cancer screening. Most also provided written explanations for their rankings and these were classified into explanatory themes. Logistic regressions using socio-demographic and attitude data predicted the type of screening chosen as first priority. The analysis revealed that many men and women did indeed assign similar priorities to the different types of screening and, even when the priorities differed, these were often justified by similar arguments relating to technical aspects of the interventions and to self-interest. However, women were far more likely than men to prioritise a type of screening from which they themselves would benefit directly and the variations in preferences and explanations between the sexes occurred primarily because of differences in other-regarding attitudes. The bias towards screening of females was driven by women's greater worries about the disease in question and by men's “benevolent sexism” with respect to women's wellbeing.     

考虑项目关联关系的R&D网络风险传播建模与仿真

考虑同一企业内不同项目间关联关系的情况下,基于研发项目的数量将研发网络分为不同子网络。借鉴SIS模型构建了R&D网络风险传播动力学模型,并对该模型进行了数理解析与仿真分析。结果表明：R&D网络风险传播阈值是由不同子网络的拓扑结构以及企业内不同项目间的关联程度共同决定的;考虑项目关联的研发网络风险传播阈值小于不考虑关联关系的研发网络风险传播阈值;项目间关联关系的企业个数及项目间的关联程度对风险在不同子网络上的传播速度与传播范围均具有影响。该研究成果对于提高研发网络抗风险能力具有重要意义。     

远海岛礁冷链物流体系优化模型及其集成算法

为研究远海岛礁渔获冷链物流的运行方案,本文选择了岛礁冷库储存中转和海上直接收购等两种不同的渔获运输方式,同时考虑了中转冷库选址、中转运输和直接收购船舶配备、航线配置、在渔场停留时间和渔获运输损耗等诸多复杂问题,构建了远海冷链物流体系组合优化模型。求解时,提出了一种以遗传算法（GA）为外部框架,以改进的模拟植物生长算法（PGSA）为内部模块,通过信息传递实现内外连接交互的集成算法。最后,以我国南海远海岛礁及渔场为实例进行分析,取得了较好效果。通过与传统遗传算法对比,本文优化结果、稳定性及计算速度等均有明显提高,从而证明了本文所建模型与算法的合理性和有效性。     

The Use of Simulation to Reduce the Domain of “Black Swans” with Application to Hurricane Impacts to Power Systems

Recently, the concept of black swans has gained increased attention in the fields of risk assessment and risk management. Different types of black swans have been suggested, distinguishing between unknown unknowns (nothing in the past can convincingly point to its occurrence), unknown knowns (known to some, but not to relevant analysts), or known knowns where the probability of occurrence is judged as negligible. Traditional risk assessments have been questioned, as their standard probabilistic methods may not be capable of predicting or even identifying these rare and extreme events, thus creating a source of possible black swans. In this article, we show how a simulation model can be used to identify previously unknown potentially extreme events that if not identified and treated could occur as black swans. We show that by manipulating a verified and validated model used to predict the impacts of hazards on a system of interest, we can identify hazard conditions not previously experienced that could lead to impacts much larger than any previous level of impact. This makes these potential black swan events known and allows risk managers to more fully consider them. We demonstrate this method using a model developed to evaluate the effect of hurricanes on energy systems in the United States; we identify hurricanes with potentially extreme impacts, storms well beyond what the historic record suggests is possible in terms of impacts.     

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.