Understanding machine learning-based forecasting methods: A decomposition framework and research opportunities

Machine learning (ML) methods are gaining popularity in the forecasting field, as they have shown strong empirical performance in the recent M4 and M5 competitions, as well as in several Kaggle competitions. However, understanding why and how these methods work well for forecasting is still at a very early stage, partly due to their complexity. In this paper, I present a framework for regression-based ML that provides researchers with a common language and abstraction to aid in their study. To demonstrate the utility of the framework, I show how it can be used to map and compare ML methods used in the M5 Uncertainty competition. I then describe how the framework can be used together with ablation testing to systematically study their performance. Lastly, I use the framework to provide an overview of the solution space in regression-based ML forecasting, identifying areas for further research.     

A fast and scalable ensemble of global models with long memory and data partitioning for the M5 forecasting competition

This work presents key insights on the model development strategies used in our cross-learning-based retail demand forecast framework. The proposed framework outperforms state-of-the-art univariate models in the time series forecasting literature. It has achieved 17th position in the accuracy track of the M5 forecasting competition, which is among the top 1% of solutions.     

辣木树不同部位γ-氨基丁酸的HPLC分析

用高效液相色谱法检测了辣木树不同部位γ-氨基丁酸的含量,结果表明:每100 g样品中,辣木果实、叶、果皮、花、根、籽及茎中的GABA含量分别为378.39、275.64、194.15、96.42、69.14、27.32 mg与0.00 mg。由此可见,辣木树中富含GABA,且幼嫩部位含量较高,本研究可为辣木产品的质量评价、开发利用以及辣木树中GABA的合成部位与运输情况研究提供理论参考。     

曲酸摇床发酵与发酵罐发酵工艺参数的比较

通过单因素及正交实验,比较高产米曲霉在摇床培养与发酵罐培养中生产曲酸的发酵工艺参数。结果表明,在温度为23℃、发酵天数为9 d、接种量为25%、菌龄为96 h时,摇床发酵的曲酸产量达到最高,最高可达98.6 g·L-1;而发酵罐培养的最佳工艺条件是温度23℃、发酵天数4 d,接种量15%,菌龄96 h时,此条件下曲酸产量可达到99.6 g·L-1。本研究为工业化生产曲酸提了供理论依据。     

Neighbors and friends: How do European political parties respond to globalization?

This paper explores the relationship between globalization and party positions accounting for potential differences between left- and right-wing parties. The analysis is based on a panel model of 36 political parties in 18 Western European countries between 1970 and 2015. We find that right-wing parties move leftward in response to globalization, while left-wing parties do not alter their position. Additionally, we find that ideological party positions are affected by foreign parties’ positions of the same ideological bloc. These findings indicate that there is a convergence towards the left of the political spectrum due to right-wing party moderation.     

一种基于SIR结构的三通带带通滤波器

应用双指耦合结构和枝节加载谐振器（Stub-loaded Resonator，SLR）实现了一款基于阶梯阻抗谐振器（Stepped Impedance Resonator，SIR）的滤波器。该滤波器具有3个通带，带外抑制较好，工作频段提高。通过调整阻抗比可调节第二、三通带的谐振频率；SLR结构能够增加通带数量；SLR结构和双指耦合结构均能改善滤波器的S参数。HFSS软件仿真表明，3个通带的中心频率分别为3.5 GHz、6.6 GHz、9.2 GHz，对应的分数带宽分别为5.7%、3%、2%，S11分别为-18 dB、-22 dB、-24 dB，通带内的S21分别为-1.8 dB、-1 dB、-1 dB。电路的测量结果与仿真结果较为吻合。该滤波器在5G通信的低频段具有应用前景。     

Diagnosing a healthcare cybersecurity crisis: The impact of IoMT advancements and 5G

Internet of Medical Things (IoMT) technology remains in early stages of adoption, but advancements and breakthroughs are quickly moving this process forward. There is a critical need for cybersecurity to be a priority in the development of these new tools, alongside design and utility. Given the rapid pace and potential magnitude of the coming advancements in IoMT, if privacy and security risks are neglected, a significant crisis could emerge in the form of more frequent cybersecurity breaches. This article examines the market opportunities and risks associated with IoMT and outlines a plan for proactively mitigating concerns and providing a platform to foster growth, to modify attitudes and behaviors, and to continue to build consumer confidence in the overall health system without sacrificing security.     

M5 competition uncertainty: Overdispersion,distributional forecasting,GAMLSS, and beyond

The M5 competition uncertainty track aims for probabilistic forecasting of sales of thousands of Walmart retail goods. We show that the M5 competition data face strong overdispersion and sporadic demand, especially zero demand. We discuss modeling issues concerning adequate probabilistic forecasting of such count data processes. Unfortunately, the majority of popular prediction methods used in the M5 competition (e.g. lightgbm and xgboost GBMs) fail to address the data characteristics, due to the considered objective functions. Distributional forecasting provides a suitable modeling approach to overcome those problems. The GAMLSS framework allows for flexible probabilistic forecasting using low-dimensional distributions. We illustrate how the GAMLSS approach can be applied to M5 competition data by modeling the location and scale parameters of various distributions, e.g. the negative binomial distribution. Finally, we discuss software packages for distributional modeling and their drawbacks, like the R package gamlss with its package extensions, and (deep) distributional forecasting libraries such as TensorFlow Probability.     

What motivates customers to shop in smart shops? The impacts of smart technology and technology readiness

With the rapid diffusion of smart technologies, a new retail mode, the smart shop, has received increasing attention from both academia and practitioners in the 5G era. However, previous studies on smart shops have largely focused on the effects of smart technologies on technology adoption rather than customer shopping behaviors. To fill this gap, this study applies the hedonic information systems acceptance model (HISAM) to identify the utilitarian and hedonic motivations affecting consumer shopping intention. In addition, we characterize a second-order formative construct, technology readiness, as a technology-related personality to test its moderating role in the research model in the marketing context when consumer behaviors may differ due to individual characteristics. Using structural equation modeling (SEM), we applied SmartPLS 3.2.8 to analyze 298 valid samples. The results show that perceived ease of use significantly affects perceived usefulness and perceived enjoyment, in turn, these three factors directly influence shopping intention. Additionally, perceived ease of use will have a stronger impact on perceived usefulness and shopping intention when the customer has a high level of technology readiness. Finally, theoretical and practical suggestions and future research directions are also discussed.     

Public policy targets in EU broadband markets: The role of technological neutrality

The European Commission has recently sought to substantially revise how it regulates the telecommunication industry, with a key goal being to incentivise investment in high-speed broadband networks. Ambitious goals to incentivise investment in high-speed broadband networks have been set across the European Union, initially in the ‘Digital Agenda for Europe’ and more recently in its ‘Gigabit strategy’. These goals reflect the view of many that there are widespread and significant socio-economic benefits associated with broadband. Our analysis explores the consequence of target setting at a European level, in terms of encouraging investment and picking which technology should be adopted within the context of technological neutrality. We demonstrate that while public policy targets might implicitly favour specific technologies, especially when gigabit targets are defined, the technological choices that occur within individual Member States are shaped by the complex and dynamic interaction between a series of path dependencies that may vary significantly across as well as within Member States. Adopting an ecosystem perspective, we propose a conceptual framework that identifies the key factors associated with technological neutrality and informs a rational decision-making process.