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.     

Effects of price sorting display on extreme option choice aversion: The role of ease of comparison in multiple option displays

Due to the popularity of online travel agency (OTA) booking platforms, OTAs display diverse information, including features of products, on their websites. Based on choice architecture literature, this study aimed to examine the effect of product quality and price sorting (vs. non-sorting) on extreme option choice aversion and identify the moderating effect of displays that made it difficult to read information. The results of a series of four experimental studies (total n = 2838 online panel members) demonstrated that the tendency to choose the non-extreme or middle-attributed options was stronger when quality and price were displayed sorted (vs. non-sorted). It was theorized that easy comparison of multiple options leads to decision-making. The positive effect of quality and price sorting on extreme option choice aversion was significantly reduced when customers had difficulty reading the displayed information.     

西部地区财政科技支出的经济增长效应

科学技术是第一生产力，科技投入无论是在推动经济增长还是促进社会发展方面，都有着举足轻重的重要作用。通过协整检验、格兰杰因果检验等方法，研究2006年-2020年西部地区财政科技支出对区域经济增长的作用。实证结果表明，西部地区的财政科技支出与经济增长之间存在着长期的均衡关系。因此，应进一步加大及合理安排政府财政科技资金的投入，加强资金监管，提高资金使用效率，更好地促进经济发展与社会进步。     

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.     

考虑声誉效应的科技服务业与制造企业合作创新行为策略分析

采用博弈论的方法，考虑声誉效应对博弈双方效用、策略的影响，建立信号博弈模型，并分析精炼贝叶斯均衡，证明声誉水平、对声誉的态度以及技术市场价值的高低是影响建立合作关系的重要因素。探讨双方建立合作关系的行为策略，提出实现合作达成双赢的建议。     

The daughter effect: do CEOs with daughters hire more women to their board?

Using a sample of S&P 100 firms, we find that CEOs with a daughter are more likely to hire new women to their board of directors than CEOs without a daughter. Our results provide additional evidence that parents’ attitudes and actions are affected by the gender of their children and that the effect is strong enough to influence important decisions at large corporations.     

The antecedents of tourists’ gaming spend: does the brand prestige matter?

This study endeavors to empirically analyse the influential effect among brand prestige, service quality, casino visitor satisfaction and spend. Data from 304 tourists who had travelled to Macau and had participated in gambling activities were used and structural equation modelling and multi-group analysis were employed in hypotheses testing. The findings demonstrated that both brand prestige and perceived service quality are determinants of satisfaction and gaming spend, and that brand prestige also affects service quality positively. Furthermore, past experience moderates the service quality and visitors’ satisfaction relationship, also service quality and visitors’ gaming spend relationship.     

Pricing used books on Amazon.com: a spatial approach to price dispersion

This paper estimates a spatial autoregressive (SAR) model of price dispersion using publicly available internet bookselling data. It uses a semiparametric adaptive estimator that does not require the usual Gaussian assumption of maximum likelihood (ML) estimators. The results suggest that both price competition and seller heterogeneity are key drivers of the observed price dispersion. The paper finds that sellers with large sales volume, newly established sellers and US mainland states-based sellers tend to price lower. The identified significant spatial interaction is evidence of spatial price competition. Controlling for everything else, a seller asks a lower price when large sellers charge relatively high prices, which is also evidence of price-based selling and undercutting.     

The signaling effects of education in the online lending market: Evidence from China

In this study, we use data from an online lending platform named Xinxindai in China to empirically study the signaling effects of education for the default risk of borrowers. Three dependent variables are created, namely, the probability of default, overdue payments and overdue amount, and probit models, count models and Tobit models are employed correspondingly. The number of universities in the “211 Project” of China at the city level is employed as the instrumental variable. The empirical evidence shows that education generally plays a strong signaling role in the identification of borrowers’ default risk in China. The negative marginal effect of education declines as borrowing times increase and as the marketization of regions deepens. This study helps to fill an important gap in the existing literature. Platforms and lenders can use educational level for reference in identifying the default risk of borrowers.     

Institutions and the financial development–economic growth nexus in sub-Saharan Africa

Using a sample splitting approach that does not impose an exogenous quadratic term, we examine the effect of financial development on economic growth in sub-Saharan Africa by allowing the link to be mediated by the level of institutions. Our findings reveal a disproportionate growth-enhancing effect of finance, given countries’ distinct level of institutional quality. More specifically, when the International Country Risk Guide-based measure of institutions is used as the threshold variable, below the optimal level of institutional quality, financial development does not significantly promote economic growth. For countries with institutional quality above the threshold, higher finance is associated with growth. However, when institutions are measured by World Governance Indicators proxy, we find a significant effect of financial development, irrespective of whether a country is below or above the threshold. Interestingly, the growth-enhancing effect of finance is greater for low-institution countries relative to high-institution countries. Thus, through its ability to provide some crucial roles, the well-developed financial sector may also perform the function of sound institutions in influencing economic growth.