Multivariate dynamic modeling for Bayesian forecasting of business revenue

Forecasting enterprise-wide revenue is critical to many companies and presents several challenges and opportunities for significant business impact. This case study is based on model developments to address these challenges for forecasting in a large-scale retail company. Focused on multivariate revenue forecasting across collections of supermarkets and product categories, hierarchical dynamic models are natural: these are able to couple revenue streams in an integrated forecasting model, while allowing conditional decoupling to enable relevant and sensitive analysis together with scalable computation. Structured models exploit multi-scale modeling to cascade information on price and promotion activities as predictors relevant across categories and groups of stores. With a context-relevant focus on forecasting revenue 12 weeks ahead, the study highlights product categories that benefit from multi-scale information, defines insights into when, how, and why multivariate models improve forecast accuracy, and shows how cross-category dependencies can relate to promotion decisions in one category impacting others. Bayesian modeling developments underlying the case study are accessible in custom code for interested readers.     

Epitaxy of Binary Compounds and Alloys

The present work aims at constructing a theoretical framework within which to address the issues of morphological instabilities (one-dimensional step bunching and two-dimensional step meandering), alloying, and phase segregation in binary systems in the context of (physical or chemical) vapor deposition. The length scale of interest, although nanoscopic, is sufficiently large that the steps on a vicinal surface can be viewed as smooth curves and, correspondingly, the theory is a continuum one. In a departure from theories inaugurated by Burton, Cabrera, and Frank [The growth of crystals and the equilibrium structure of their surfaces. Phil. Trans. Roy. Soc. A 243 (1951) 299–358] the steps are endowed with a thermodynamic structure whose main ingredients are a step free-energy density and edge species chemical potentials. Moreover, crystal anisotropy, with its altering of the dynamics of steps and the associated morphological instabilities, is accounted for – in a manner consistent with the second law – both in the thermodynamic and kinetic properties of terraces and, more importantly, of steps. Additionally, in contrast with most of the literature on the subject (cf. [J. Krug, Introduction to step dynamics and step instabilities. In: A. Voigt (ed.) Multiscale Modeling in Epitaxial Growth. Birkhäusser, Berlin (2005)]), adsorption–desorption along the steps, bulk atomic diffusion, and chemical reactions (both on the terraces and along the step edges) are incorporated and coupled to the other mechanisms, e.g., terrace adatom diffusion and step attachment–detachment kinetics, whose interplay governs the evolution of steps on vicinal surfaces. Importantly, aided by the concept of configurational forces for which a separate balance law is postulated Configurational Forces as Basic Concepts of Continuum Physics. Springer, Berlin Heidelberg New York (2000)]), the proposed theory allows the steps to evolve away from local equilibrium thus contributing to a general treatment of the dynamics of steps. Finally, a specialization to the epitaxy of binary compounds and alloys is afforded, yielding a generalization of the classical Gibbs–Thomson relation in the former and novel evolution equations for an individual step in the latter.     

手机广告链中移动运营商的努力动机研究与仿真

建立了由单手机广告主、手机广告代理商和移动运营商构成的三级广告产业链之间的基于协作的收益共享契约模型,分析了在以移动运营商为主导的手机广告链中,移动运营商的努力动机受收益共享和努力成本补偿参数的影响.采用量化分析和模拟仿真结合的方法,得出契约参数只有在满足一定条件下,移动运营商所做出的最优努力水平投入决策,将使得整个手机产业链收益最大化,实现多方共赢.     

澳门博采收入增长影响因素分析

通过对2005年1月至2013年7月份的澳门博采毛收入数据进行季节分析,显示澳门博采毛收入有显著的中国节假日特征.其次分析入境澳门旅客数、酒店住客数、劳动力参与率、消费指数以及突发事件是否是澳门博采毛收入迅速增长的原因,通过计量经济学分析,得到酒店住客数、劳动力参与率、中国的节假日三个因素对澳门博采毛收入有显著的正向影响关系,而2008年世界金融危机对博采毛收入有显著的负的影响.最后结合实际给出了提高博采收入的政策建议.     

Allocating fixed costs and common revenue via data envelopment analysis

An issue of considerable importance involves the allocation of fixed costs or common revenue among a set of competing entities in an equitable way. Based on the data envelopment analysis (DEA) theory, this paper proposes new methods for (i) allocating fixed costs to decision making units (DMUs) and (ii) distributing common revenue among DMUs, in such a way that the relative efficiencies of all DMUs remain unchanged and the allocations should reflect the relative efficiencies and the input-output scales of individual DMUs. To illustrate our methods, numerical results for an example are described in this paper.     

基于Shapley值法的三级闭环供应链收益分配模型

以合作博弈为基本研究方法,构建了由制造商、第三方物流和零售商组成的三级闭环供应链收益分配模型,不同合作方式下各合作联盟收益的特征函数的计算结果表明:在采取合作时,各合作联盟的收益均大于或等于各成员在不合作时各自的收益之和,文章验证了在三级闭环供应链收益分配模型中应用Shapley值法的可行性,说明构建的收益分配模型具有普遍适用性.然后运用Shapley值法求得三级闭环供应链收益分配的合理均衡解,最后结合算例进行了分析论证.     

非对称信息和模糊需求下的闭环供应链协调决策研究

在模糊需求和非对称销售价格信息下,研究了由单一制造商和零售商组成的两级闭环供应链系统的协调问题。通过建立不确定环境下的集中决策和收益共享-费用共担契约决策模型,分别得到了不同模型下闭环供应链系统的最优策略。并进一步通过算例分析了契约下供应链各方收益随模糊需求和价格的变化情况,以及该契约对整个系统决策效率的影响。结果表明:收益共享-费用共担契约能有效实现模糊闭环供应链系统的协调,实现闭环供应链系统决策的最优化;其次,需求和零售价格的不确定性程度越高,对零售商的最优订购量、供应链成员及系统利润的影响也越大。     

二级物流服务供应链的契约协调机制研究

研究一个物流服务集成商和一个物流服务提供商所组成的二级物流服务供应链的系统协调问题。分别建立了集中控制模型、批发价格契约下的Stackelberg主从协调模型、收益共享契约下的协调模型,得到不同协调方法下的最优解所满足的条件。通过模型间的对比,证明了收益共享契约可使得该物流服务供应链达到整体最优的状态,相对于批发价格契约,实现了Pareto改进。最后,用数值试验的方式进一步验证了收益共享契约在协调物流服务供应链运作的有效性。     

基于战略顾客行为的最优定价与容量选择模型

研究了当市场中同时存在战略顾客和短视顾客时零售商的最优定价与容量选择问题。零售商在正常销售阶段和出清销售阶段制定不同的销售价格,同时通过容量选择影响战略顾客的购买行为,而战略顾客则根据零售商的定价和容量选择确定最优购买时机。分别分析了零售商在无限容量时的定价决策、固定价格时的容量选择、固定容量时的定价决策以及有限容量下的定价与容量选择四种情形。研究结果表明,零售商在无容量限制时的最优定价决策是制定两阶段定价策略,在固定价格时的最优容量选择依赖于模型的参数,而当零售商的容量固定时,部分满足出清销售阶段的顾客需求始终优于完全满足出清销售阶段的顾客需求。     

网络团购的定价与收益协调策略研究

如何实现团购网站和团购参与商家之间的双赢是我国网络团购目前面临的一大挑战性问题。结合我国网络团购的特点,构建团购网站和商家之间定价与收益分配的博弈模型,针对博弈结果运用纳什谈判模型给出协调策略。最后,通过数值算例分析主要参数的变化对价格与收益的影响。结果表明,敏感市场的规模和团购网站的广告效应越大、商品的单位成本越小、开展网络团购对商家和团购网站越有利。