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.     

Perseverance and suspense in tug-of-war

We study a tug-of-war game between two players using the lottery contest success function (CSF) and a quadratic cost (of effort) function. We construct a pure strategy symmetric Markov perfect equilibrium of this game, show that it is unique, and provide closed-form solutions for equilibrium strategies and values. In stark contrast to a model of tug-of-war with an all-pay auction CSF, players exert positive efforts until the very last battle in this equilibrium. We deliver a set of empirically appealing results on effort dynamics.     

水生态系统服务供需视角下的水网乡村 适应性规划策略——以吴江张鸭荡片区为例

苏南乡村地区一直是中国乡村建设的先行区域，特殊的纵横交错的水网结构构成了一个错综复杂的系统，呈现出其他地区不具有的复杂性和生态特殊性。随着城镇的扩张，乡村的发展建设使其水网空间的平衡发展面临极大的挑战。传统单一静态的规划方法逐渐显示出无法适应经济、社会等发展要求的缺陷。苏南乡村地区以水为核心，从水生态系统服务供需关系的视角下探究水网乡村的适应性规划策略更加适应当前的乡村现状和需求。以传统水网乡村空间形态转译为基础，建立水生态系统供需服务评估体系，在评估水生态系统服务供需能力的基础上，分析供需分异模式及供需矛盾。从构建乡村水域空间生态格局、乡村水域空间功能分区规划，以及乡村水域空间多情景预判3个方面，提出苏南水网乡村的适应性规划策略，并为水网乡村的生态实践提出新思路。     

非国有大股东对国企超额雇员的影响:成本效应还是激励效应

基于竞争理论的国企改革思路强调减少政策性负担摊派,基于产权理论的改革思路侧重于引入非国有资本,以解决产权不清晰与所有者缺位问题.本文尝试融合两种理论,以国企超额雇员问题为研究对象,从非国有大股东视角探究混合所有制改革成效.研究发现:非国有大股东能够显著降低国企超额雇员,减弱超额雇员对公司价值的损害作用,且该治理作用会随非国有大股东相对力量的提高而增强;从作用机制来看,该治理作用不仅表现为劳动力成本尤其是员工薪酬的节约,还表现为内部薪酬差距和管理层薪酬业绩敏感性的提高,兼具成本效应与激励效应;该作用在地区就业压力较小、劳动密集度较低以及劳动保护水平更低的环境下更为显著.     

Optimal targeting of latent tourism demand segments

This paper identifies for the first time the optimal target markets employing the latent tourism demand expenditure, a novel concept in tourism literature. The study quantifies latent tourism demand between each pair of origin-destination through distinguishing by type of tourism and seasonality. It works with market shares that are estimated via a fractional regression model. Moreover, latent demand is clustered using a market segmentation approach based on a latent class regression. Finally, the optimal target markets are chosen depending on the expected latent tourism expenditure. The result has clear policy implications in terms of which markets are promoted, the optimal channels of communication and the maximum budget for each marketing campaign.     

Is the system reliability profitable for retailing and consumer service of a dynamical system under cross-price elasticity of demand?

Retailing strategy is one of the most crucial factors for industries. A proper retailing strategy can help to enhance consumer service and increase the industry's profit. An improved approach to retailing is suggested in this research to deliver superior customer service while maximizing profits in a dynamic system. The study analyzes a retailing strategy for a demand with cross-price elasticity upon the retail price. A product's cross-price elasticity and the system reliability are critical factors in retailing. Understanding the cross-price elasticity of demand between products helps retailers to make pricing decisions that maximize profits by maintaining demand. Imperfect products are produced due to an imperfect production system. The imperfect ones must be adjusted with some costs to make them perfect for better retailing. The system failure rate is crucial for retailing under cross-price elasticity of demand patterns. Production system reliability, cross-price elasticity of demand, and consumer service are all essential factors that can impact a company's success in the market. The production rate is considered time- and system failure rate-dependent. Contradictory to the literature, a dynamical system is proposed for improved retail management, which is solved using the Euler-Lagrange theory. Finally, one can achieve the expected maximum profit for this retail system with optimum selling prices for different products by reducing the system failure rate. Some numerical illustrations with graphical representations are provided to validate the current study. Numerical examples show that applying cross-price elasticity of demand for more than two identical products provides 35% more profit for the retail industry than a single type of product.     

Cryptocurrencies and Cagan’s model of hyperinflation

The drivers of the prices of Bitcoin and Ethereum are studied within a framework based on Cagan’s model of hyperinflation. In the model, the prices of the cryptocurrencies are driven by stochastic adoption and velocity shocks as well as endogenous expectations of future prices. The model is estimated with data for prices, transaction volumes, and money supplies. A majority of price fluctuations in both currencies can be attributed to shocks in adoption, velocity shocks are much less important. The money demand sensitivity to expected price changes is estimated to be larger for Bitcoin than for Ethereum, and both have higher sensitivity than fiat currencies during episodes of hyperinflation.     

Advertisement policy for dual-channel within emissions-controlled flexible production system

Nowadays, industrial firms are very much careful to build a green environment by reducing carbon emissions. The government imposes some rules and regulations to provide a better eco-friendly environment. In this study, the cap-and-trade mechanism has been considered in a production model to control the carbon emissions rate. The manufacturers invest in advanced green technology to reduce per unit emissions. As online and offline selling is crucial to any industry for increasing customer demand, the manufacturers sell their products by dual-channel and advertise their products by online channel to make more popular of their products. Keeping these in mind, a sustainable flexible production model with single-type substitutable product production is considered here by imposing a cap-and-tax policy, investing in green technology, and advertising for products. This model is divided into two cases: with and without investment in green technology. The demand of each manufacturer depends upon an online selling price, an offline selling price, and an online advertisement of the product. A classical optimization technique helps to get the optimum strategies for the online selling price, offline selling price, advertisement investment, green technology, cycle time, and production rate. From the numerical study, it is proved that the industry gets 6.97% more profit in the case of green technology investment and the proposed study gives 5.74% more profit than the traditional production system. Sensitivity analysis and managerial insights are performed.     

Debt Overhang and Deleveraging in the US Household Sector: Gauging the Impact on Consumption

Using a novel data set for the U.S. states, this paper examines whether household debt and the protracted debt deleveraging help explain the dismal performance of U.S. consumption since 2007 in the aftermath of the housing bubble. By separating the concepts of deleveraging and debt overhang—a flow and a stock effect—we find that excessive indebtedness exerted a meaningful drag on consumption over and beyond wealth and income effects. The overall effect, however, is modest—‐around one sixth of the slowdown in consumption between 2000–06 and 2007–12—and mostly driven by states with particularly large imbalances in their household sector. This might be indicative of non‐linearities, whereby indebtedness begins to bite only when misalignments from sustainable debt dynamics become excessive.     

Predictive regressions under asymmetric loss: Factor augmentation and model selection

This paper discusses the specifics of forecasting using factor-augmented predictive regressions under general loss functions. In line with the literature, we employ principal component analysis to extract factors from the set of predictors. In addition, we also extract information on the volatility of the series to be predicted, since the volatility is forecast-relevant under non-quadratic loss functions. We ensure asymptotic unbiasedness of the forecasts under the relevant loss by estimating the predictive regression through the minimization of the in-sample average loss. Finally, we select the most promising predictors for the series to be forecast by employing an information criterion that is tailored to the relevant loss. Using a large monthly data set for the US economy, we assess the proposed adjustments in a pseudo out-of-sample forecasting exercise for various variables. As expected, the use of estimation under the relevant loss is found to be effective. Using an additional volatility proxy as the predictor and conducting model selection that is tailored to the relevant loss function enhances the forecast performance significantly.