Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WC20

3 - The Sharing Economy and Housing Affordability: Evidence from Airbnb Davide Proserpio, University of Southern California, 701 Exposition Blvd, HOH 332, Los Angeles, CA, 90034, United States We assess the impact of home-sharing on residential house prices and rents. Using a dataset of Airbnb listings from the entire United States and an instrumental variables estimation strategy, we find that a 1% increase in Airbnb listings leads to a 0.018% increase in rents and a 0.026% increase in house prices at the median owner-occupancy rate zip code. The effect is moderated by the share of owner- occupiers, a result consistent with absentee landlords reallocating their homes from the long-term rental market to the short-term rental market. A simple model rationalizes these findings. 4 - The Effect of User Generated Content on Hotel Demand under a Competitive Framework Sanghoon Cho, University of South Carolina, 1014 Greene Street, Columbia, SC, United States, Pelin Pekgun, Ram Janakiraman We investigate the impact of user generated content on hotel performance as captured by actual hotel bookings. We propose and estimate a consumer learning model that focuses on the effect of review sentiment on hotel demand taking into account the effect of competition and hotel prices. Drawing on prospect theory, we cast the competitive perception spillover effects in the form of gains and losses in review sentiment, and find that the relative negative sentiment has more influence on bookings than the relative positive sentiment. n WC22 North Bldg 130 Practice – Finance Theory & Financial Engineering I Contributed Session Chair: Min Dai, National University of Singapore, Dept of Math, Singapore, 119076 1 - Investment and Financing for Cash Flow Discounted With Group Diversity Zhaojun Yang, Southern University of Science and Technology, 1088 Xueyuan Avenue, Fourth Building of the Faculty Apartment, Shenzhen, 518055, China We consider a firm’s investment and financing decisions made by a group of which each individual may utilize different discount rates to price cash flow generated by the investment project. We show that a higher degree of decreasing impatience or a greater group diversity increases the project value, accelerates investment and postpones default. Both the value of the investment option and the optimal leverage increase with the degree of the decreasing impatience and the group diversity. The inefficiency from asset substitution increases but it from debt overhang decreases with the degree of the decreasing impatience and the group diversity. Our predictions are documented by empirical evidences. 2 - Making the Most Out of Market Forecasts Based on Linear Regression Hamed Khaledi, PhD Student, Broad Graduate School of Management, 632 Bogue Street, North Business College Complex, Room N203, East Lansing, MI, 48825, United States This paper presents a mathematical model to make optimal trading decisions using forecasts made by multivariate regression. Given that these results are uncertain, the model maximizes the expected profit from opening and closing trade positions. To this end, a dynamic programming approach is employed. We first find the optimal take profit levels associated with buy and sell positions. Then we decide to buy, sell or wait, based on the maximum expected profit in each case. The approach is applied to find the optimal strategy for trading GBPUSD rate based on a multivariate regression model fitted to the historical daily data in the FOREX Market. 3 - The Design of a Global Supply Chain by Integrating Operational and Financial Strategies Mengyue Wang, PHD, Tsinghua University, Beijing, China, Hongxuan Huang In this paper, a scenario-based mixed-integer linear programming model is proposed for designing a flexible capital-constrained global supply chain (CCGSC) in which integrate the applying for loans with operational strategies such as constructing, or leasing facilities and equipments to meet uncertain demands and exchange rates. The research on the CCGSC also indicates that the complementary property holds for both the remaining capital of loans with higher costs and another available loan with a lower cost under certain conditions. A case study is also presented to illustrate effectiveness and efficiency of the scenario-based approach.

n WC20 North Bldg 129A Matching Markets Sponsored: Revenue Management & Pricing Sponsored Session Chair: Ozge Sahin, Johns Hopkins University, Brooklyn, NY, 11222, United States 1 - Information Acquisition Costs of Matching Markets Irene Yuan Lo, Columbia University, New York, NY, 10025-6186, United States Centralized matching mechanisms are typically designed assuming all participants know their preferences. We study the welfare effects of costly information acquisition in a matching model where school priorities are known, but students learn their preferences at a cost. Traditional matching mechanisms that do not account for costly information acquisition can lead to information deadlocks, where it is strictly optimal for every agent to wait for others to provide information before learning their own preferences. We characterize the optimal omniscient mechanism using an admission cutoff structure and propose mechanisms that learn market information to approximate the optimal outcome. 2 - Bonus Competition in the Gig Economy Xiaoyan Liu, Cornell University, Sage Hall, Ithaca, NY, 14853, United States, Yao Cui, Li Chen Unlike traditional firms, service platforms in the gig economy need to compete for both customers and independent service providers. We study how service platforms can offer bonuses to attract service providers in a competitive setting, and investigate the implications of platforms’ bonus strategies on the stockholders of the gig economy. 3 - Nonlinear Pricing for Online Matching Markets Ozge Sahin, Johns Hopkins University, Baltimore, MD, 21231, United States Online matching platforms such as eHarmony lack a straightforward quality measure, and users’ preferences are often fluid and subjective. A large and active user base is a platform’s most precious asset. We investigate how pricing mechanisms can help manage user entry, and exit behavior to the matching platform to improve social welfare and firm revenues. n WC21 North Bldg 129B Innovative Practices in Pricing and Revenue Management Sponsored: Revenue Management & Pricing Sponsored Session Chair: Pelin Pekgun, University of South Carolina, Columbia, SC, 29205, United States Co-Chair: Ovunc Yilmaz, University of Notre Dame, South Bend, IN, 46617, United States 1 - Inventory Count-down Information Laura Wagner, NIF 501082522, Lisbon, 1200/830, Portugal, Edu Calvo, Ruomeng Cui Online retailers provide the scarcity message such as count-down inventory information to alert consumers that there are only a few units left. Collaborating with the leading flash-deal retailer in Spain, which displays the limited-inventory information for the last five inventory units, we explore how it affects consumer purchasing behavior. We use a difference-in-difference approach to examine the impact on sales, returns and net sales. We find that the scarcity message causally attracts more sales, the return rate remains the same and net sales increase. We also investigate the moderating effect of product characteristics, and how the scarcity effect varies across web and mobile channels. 2 - The Effect of Platform Selection and Underreporting Biases on Tripadvisor Ratings Saram Han, 110 Joanne Dr., Ithaca, NY, 14850, United States Cornell SC Johnson College of Business, Ithaca, NY, United States This study shows how an underreporting bias affects the shape and average of an online satisfaction rating in TripAdvisor in conjunction with a platform selection bias. We show that since expert reviewers’ expectations for service quality is hard to meet, most of the properties experience negative platform selection bias. The direction of the underreporting bias is determined by the reporting propensity that depends on how strongly the reviewers are motivated to report their satisfaction, whether to complement or complain. The bounds of the reporting propensity are determined by the reviewers’ expertise with the platform.

476