-
作者：Blanc, Sebastian M.; Setzer, Thomas
摘要：Combining forecasts is an established approach for improving forecast accuracy. So-called optimal weights (OWs) estimate combination weights by minimizing errors on past forecasts. Yet the most successful and common approach ignores all training data and assigns equal weights (EWs) to forecasts. We analyze this phenomenon by relating forecast combination to statistical learning theory, which decomposes forecast errors into three components: bias, variance, and irreducible error. In this framew...
-
作者：Alizamir, Saed; de Vericourt, Francis; Wang, Shouqiang
作者单位：Yale University; European School of Management & Technology; University of Texas System; University of Texas Dallas
摘要：The World Health Organization seeks effective ways to alert its member states about global pandemics. Motivated by this challenge, we study a public agency's problem of designing warning policies to mitigate potential disasters that occur with advance notice. The agency privately receives early information about recurring harmful events and issues warnings to induce an uninformed stakeholder to take preemptive actions. The agency's decision to issue a warning critically depends on its reputati...
-
作者：Khapko, Mariana; Zoican, Marius
作者单位：University of Toronto; University Toronto Scarborough; University of Toronto; University Toronto Mississauga
摘要：Recent regulatory and industry initiatives aim to streamline post-trade infrastructures. Does faster settlement benefit markets? We build a model of intermediated trading with imperfectly competitive securities lending. Faster settlement benefits impatient traders but increases borrowing needs. We find that flexible failure-to-deliver penalties reduce this tension, disciplining security lender competition and allowing for real-time settlement. Optimal penalties resemble put options on the lend...
-
作者：Whitaker, Amy; Kraussl, Roman
作者单位：New York University; University of Luxembourg; Stanford University
摘要：A core challenge in studying the real return on artist' work is the extreme difficulty accessing private records from when an artwork was first sold and thus relying on public auction data. In addition, artists do not typically receive proceeds after the initial sale. This paper, for the first time, uses archivally sourced primary market records to model returns on art and introduces a novel fractional equity structure for artists. We first model what would happen if the American artists Jaspe...
-
作者：Barach, Moshe A.; Golden, Joseph M.; Horton, John J.
作者单位：University of Minnesota System; University of Minnesota Twin Cities; Massachusetts Institute of Technology (MIT); National Bureau of Economic Research
摘要：Platform marketplaces can potentially steer buyers to certain sellers by recommending or guaranteeing those sellers. Money-back guarantees-which create a direct financial stake for the platform in seller performance-might be particularly effective at steering as they align buyer and platform interests in creating a good match. We report the results of an experiment in which a platform marketplace-an online labor market-guaranteed select sellers for treated buyers. The presence of a guarantee s...
-
作者：Zhang, Dennis J.; Dai, Hengchen; Dong, Lingxiu; Qi, Fangfang; Zhang, Nannan; Liu, Xiaofei; Liu, Zhongyi; Yang, Jiang
作者单位：Washington University (WUSTL); University of California System; University of California Los Angeles; Alibaba Group; Alibaba Group; Ant Group; Alibaba Group
摘要：Dynamic pricing through price promotions has been widely used by online retailers. We study how a promotion strategy, one that offers customers a discount for products in their shopping cart, affects customer behavior in the short and long term on a retailing platform. We conduct a randomized field experiment involving more than 100 million customers and 11,000 retailers with Alibaba Group, one of the world's largest retailing platform. We randomly assign eligible customers to either receive p...
-
作者：Zhang, Huanan; Chao, Xiuli; Shi, Cong
作者单位：Pennsylvania Commonwealth System of Higher Education (PCSHE); Pennsylvania State University; Pennsylvania State University - University Park; University of Michigan System; University of Michigan
摘要：We consider a periodic-review, single-product inventory system with lost sales and positive lead times under censored demand. In contrast to the classical inventory literature, we assume the firm does not know the demand distribution a priori and makes an adaptive inventory-ordering decision in each period based only on the past sales (censored demand) data. The standard performance measure is regret, which is the cost difference between a learning algorithm and the clairvoyant (full-informati...
-
作者：Chung, Kevin
作者单位：University of Wisconsin System; University of Wisconsin Madison
摘要：We study a rebate program in the credit card industry where the firm automatically processed the rebate earned by consumers with flexible and effortless redemption opportunities in future transactions. Contrary to what the normative economic theory would predict, we document that consumers frequently failed to redeem their rebate. Given this observation, we examine what may have been the primary driving force behind this pattern and the implications for the firm. We provide evidence that the p...
-
作者：Roels, Guillaume
作者单位：INSEAD Business School
摘要：Despite their idiosyncrasies, motor and cognitive learning and endurance sports training have in common that they involve repeated practice. While considerable research has been devoted to the effect of practice on performance, little is known about optimal practice strategies. In this paper, we model the practice process for both skill acquisition and retention, and optimize its profile to maximize performance on a predefined date. For skill acquisition, we find that the optimal process invol...
-
作者：Kahale, Nabil
作者单位：heSam Universite; ESCP Business School
摘要：We present a new unbiased algorithm that estimates the expected value of f(U) via Monte Carlo simulation, where U is a vector of d independent random variables, and f is a function of d variables. We assume that f does not equally depend on all its arguments. Under certain conditions, we prove that, for the same computational cost, the variance of our estimator is lower than the variance of the standard Monte Carlo estimator by a factor of order d. Our method can be used to obtain a low-varian...
