Applicability of the M5 to Forecasting at Walmart

The M5 Forecasting Competition, the fifth in the series of forecasting competitions organized by Professor Spyros Makridakis and the Makridakis Open Forecasting Center at the University of Nicosia, was an extremely successful event. This competition focused on both the accuracy and uncertainty of forecasts and leveraged actual historical sales data provided by Walmart. This has led to the M5 being a unique competition that closely parallels the difficulties and challenges associated with industrial applications of forecasting. Like its precursor the M4, many interesting ideas came from the results of the M5 competition which will continue to push forecasting in new directions.In this article we discuss four topics around the practitioners view of the application of the competition and its results to the actual problems we face. First, we examine the data provided and how it relates to common difficulties practitioners must overcome. Secondly, we review the relevance of the accuracy and uncertainty metrics associated with the competition. Third, we discuss the leading solutions and their implications to forecasting at a company like Walmart. We then close with thoughts about a future M6 competition and further enhancements that can be explored.     

基于风险价值和期权理论的Credit Metrics模型研究

Credit Metrics模型是国际金融界内流行的现代信用风险度量模型,它以风险价值VaR和期权定价思想为基础,以衡量信贷资产组合的风险价值为核心,用于识别贷款、债券等传统信贷产品的信用风险,开启了银行信用风险量化评估的先河,开创了现代信用风险度量研究的新领域,对中国商业银行的信用风险管理有一定的借鉴作用。     

The Influences of Customer Relationship Metrics on Shareholder Value: What Should Be Done and What Has Been Done?

The contribution of customer relationship (CR) metrics to shareholder value is practically unknown in research and practice. Existing studies report rather ambiguous results. We develop a proposal to identify and measure the multidimensional influences in a valid and reliable way. By comparing existing studies to each other and to our framework, we can find possible reasons for their ambiguity: First, the studies are actually not compatible while using various CR metrics and performance measures. Second, several conceptual and methodological differences also place the comparability of their results in question. The application of our framework should explain the inconsistent outcomes of existing studies and indicate a consistent and general influence of CR metrics on shareholder value.     

Models and metrics to assess humanitarian response capacity

The race to meet vital needs following sudden onset disasters leads response organizations to establish stockpiles of inventory that can be deployed immediately. These government or non-government organizations dynamically make stockpile decisions independently. Even though the value of one organization's stock deployment is contingent on others' decisions, decision makers lack evidence regarding sector capacity to assess the marginal contribution (positive or negative) of their action. To our knowledge, there exist no metrics describing the system capacity across many agents to respond to disasters. To address this gap, our analytical approach yields new humanitarian logistics metrics based on stochastic optimization models. Our study incorporates empirical data on inventory stored by various organizations in United Nations facilities and in their own warehouses to offer practical insights regarding the current humanitarian response capabilities and strategies. By repositioning inventory already deployed, the system could respond to disasters in the same expected time with a range of 7.4%–20.0% lower cost for the items in our sample.     

Designing and implementing conservation tender metrics: Twelve core considerations

In conservation tenders metrics are intended to describe the values that would result from alternate investments; they are the linkage between individual on-ground projects and tender scheme proponents. These metrics have received little attention by economists yet are the critical linchpin that defines the nature of the values that are traded in markets. In this paper we identify twelve lessons that can guide practitioners in metric design. These lessons embrace the principal biophysical changes likely to result from management change at each site and in combination, and reflect the values that society places on the outcomes from these changes. Practical application is explored through four example tenders encompassing five metrics which demonstrate substantial variation from the lessons identified. The practical consequences for cost-efficiency of investment, and for the use of conservation tenders generally, are unclear because of the practical and political difficulties in developing effective conservation tender approaches.     

Because you're worth it? Determinants of Vice Chancellor pay in the UK

Wage inequality has increased across most developed nations; this has been manifested in a wide range of organisations and sectors, with implications for well‐being and sustainability; within UK universities, this has become increasingly visible. There is increasing pressure on universities to deliver social and economic impact in an increasingly market‐driven and metric‐driven environment. In the UK context, increasing financial pressure has led to both an escalation of student fees and constrained wage growth for faculty. In contrast, most Vice Chancellors have secured substantive pay packages raising concerns that regulatory failures may be contributing to the rise. We show that Vice Chancellors use their internal power within organisations to extract a disproportionate amount of the value created by the institution. However, we encountered much diversity according to the quality of governance, highlighting the extent to which not only contextual but also internal dynamics drive wage inequality.     

Tourism’s lost leaders: Analysing gender and performance

Higher education is increasingly engaged with diversity initiatives, especially those focused on women in academic leadership, whilst there is an evolving literature across the humanities and the social, management and natural sciences, critiquing academia’s gendered hierarchies. In contrast, senior academics in the field of tourism management have largely eluded similar sustained analysis. This paper builds on recent gender-aware studies of tourism’s leading academics with three aims. Firstly, to widen evidence of gendering in tourism’s academic leadership by scrutinizing and contextualizing performance indicators, which make and mark its leaders and shape its knowledge canon. Secondly, since critique alone cannot lead to transformation, the paper seeks to ‘undo’ gender in tourism’s academy. Thirdly the paper presents interventions to accelerate academic gender equity.     

How many singularities are near and how will they disrupt human history?

This paper reviews a large number of approaches that have been used for considering technologically driven profound societal change. We agree with Vinge's suggestion for naming events that are “capable of rupturing the fabric of human history” (or leading to profound societal changes) as a “singularity”. This is a useful terminology especially since a mathematically rigorous singularity seems impossible for technological and related societal change. The overview of previous work is done within the context of a broader look at the role of technological change within human history. The review shows that a wide variety of methods have been used and almost all point to singularities in the present century particularly in the middle of the century. The diversity of the methods is reassuring about the potential robustness of these predictions. However, the subjectivity of labeling events as singularities (even well studied past events) is a concern about all of the methods and thus one must carefully pause when relying in any way on these predictions. The general lack of empirical research in this area is also a concern.Quantitative considerations (by proponents and opponents) about past singularities or future singularities often confound two types of metrics. The first type is essentially related to diffusion of technologies (or bundles of technologies) where the logistic curve is empirically well established as the proper time dependence. The second type of metric is for technological capability where hyper-exponentials are empirically well established for their time dependence. In this paper, we consider two past singularities (arguably with important enough social change to qualify) in which the basic metric is alternatively of one type or another. The globalization occurring under Portuguese leadership of maritime empire building and naval technological progress is characterized by a metric describing diffusion. The revolution in time keeping, on the other hand, is characterized by a technological capability metric. For these two cases (and thus robust to the choice of metric type), we find that:

• • 

  People undergoing profound technologically-driven societal change do not sense a singularity.

• • 

  The societal impacts depend in complicated ways on human needs, institutional variables and other more uncertain factors and thus are particularly hard to project;

• • 

  The societal impact is apparently not determined by the rate of progress on either type of metric or by projections to mathematical points with either kind of metric. This finding supports the existing concept that social change due to technology is a more holistic phenomenon than can be characterized by any technical metric.

In the final section, we use these empirical findings as the basis for exploring the possibilities for and nature of future singularities. In this we speculate that the potential for a future strong singularity based upon computational capability does not appear particularly probable but that one may already be occurring and is not fully noticed by those (us) going through it. Other possible 21st century singularities (life extension and fossil fuel elimination are two examples considered) may also be already underway rather than waiting for the predicted mid-century changes.     

企业微博营销效果评估研究

目前企业微博营销的效果评估亟待完善,在借鉴社会化媒体营销相关研究成果的基础上,结合我国企业微博运营现状,提出企业微博营销效果的评估模式与评估指标,并总结出企业微博营销效果评估的三种主要方式,试图为企业微博营销效果评估的应用提供理论与实践支撑。