Informs Annual Meeting 2017

SC30

INFORMS Houston – 2017

SC29

4 - Corporate Prediction Market with Over-Optimistic Participants Jingchuan Pu, University of Florida, 2300 SW 43rd Str, R1, Gainesville, FL, 32607, United States, jingchuan@ufl.edu, Liangfei Qiu Recently, corporate prediction market has been growingly popular for the decision making. Even though it has an important role to integrate participates’ information, the corporate prediction market is inevitably affected by the participates’ irrational thinking. Over-optimism is one of well-known irrationality in the corporation. In this study, we investigate the effect of the information precision and social interactions on the prediction market performance, when there is systematical over-optimism among the market participants. 5 - Quantifying the Effect of Social Media On Crowdfunding Campaigns Mahdi Moqri, University of Florida, 725 NW. 13th St. 3410, Gainesville, FL, 32601, United States, mahdi.moqri@warrington.ufl.edu Using a unique dataset from 590 crowdfunding campaigns observed over 4 months, we examine to what extent, and how quickly online WOM affect the rate of contributions. In addition, we explore the effect of different phases of fundraising and of the coverage of the campaigns in online news websites. Invited: Auctions Invited Session Chair: Robert Day, University of Connecticut, Storrs, CT, 06269-1041, United States, Bob.Day@business.uconn.edu 1 - The Role of Optimization in the Incentive Auction Karla L. Hoffman, George Mason University, System Eng and Operations Research Dept, 4400 University Drive Mailstop 4a6, Fairfax, VA, 22030, United States, khoffman@gmu.edu, Brian Smith, James Andrew Costa, Rudy K.Sultana, Anthony Coudert The FCC completed the Broadcast Incentive Auction on March 29, 2017. This auction reassigned television stations onto fewer channels in the United States and Canada and repurposed a significant amount of spectrum for wireless use. Optimization played a vital role in the success of the auction. This talk will reflect on the OR activities that took place before, during and after the auction. We will highlight some of the results from the auction and some of the lessons learned. 2 - Watershed Market Design: the Case of the Central Platte Groundwater Exchange David McAdams, Duke University, 100 Fuqua Drive, Durham, NC, 27708, United States, david.mcadams@duke.edu Groundwater and surface water are connected hydrologically, but limited options are available for groundwater and surface water users to contract on water use. I propose a relatively simple watershed-market design that generates location- specific water prices and respects regulatory constraints on water flows and water use. This market design was adopted by farmers in the Central Platte Natural Resources District in Nebraska in 2016. 3 - Bidding and Learning in Repeated Auctions Karti Puranam, LaSalle University, Philadelphia, PA, 19141, United States, kartys.here@gmail.com, Michael Katehakis We consider the problem of a firm that procures substitutable items in a sequence of auctions by bidding against the “market.” The firm and the “market” learn from each winning bid. We study bidding strategies for the firm when the objective of the firm is to maximize long run discounted profit. 4 - Supplier Versus Supply Base Development Karca Duru Aral, Syracuse University, 721 University Avenue, Syracuse, NY, 13244, United States, kdaralwa@syr.edu A buyer may choose to sign a long-term contract with an incumbent to encourage cost reduction (which consequently strengthens the incumbent’s monopoly). The buyer can instead choose to invest in developing entrants to challenge the incumbent in a competitive-bid procurement auction - both of which are risky options. We analyze when the buyer should break the incumbent’s monopoly and characterize how this decision changes with the buyer’s business environment. SC28 350D Auction Applications

350E Analyzing Social Networks and Social Media for Healthcare Sponsored: Artificial Intelligence Sponsored Session Chair: Xi Wang, University of Iowa, Iowa City, IA, 52246, United States, xi-wang-1@uiowa.edu 1 - Mining User-generated Data in OHC for Drug Related Information Zhao Mengnan, PhD Candidate, Drexel University, Philadelphia, PA, United States, mz438@drexel.edu A variety of data resources has been used to identify healthcare-related information, mainly including biomedical database, medical literature, electronic health record (EHR), and clinical note. While these data sources are informative, they are mainly contributed by healthcare professionals. None of them capture the information offered by health consumers. Meanwhile, the popularity of Web 2.0 not only breeds the various online social media sites like Facebook and Twitter, but also fosters online health communities (OHCs). OHCs have been widely used for exchanging health information between health consumers. We Prior studies have suggested that widely publicized celebrity suicides often lead to an increase in copy-cat suicides. This is also known as the Werther effect. Given the prevalent use of social media, social-media based conversations after the suicide of a celebrity create significant buzz and generate publicity. However, whether the Werther effect can be caused by social-media conversations on celebrity suicides remains an empirical question. In this study, we test this effect by examining whether and how social media impacts copy-cat suicides using a data set collected from Twitter. 3 - The Impact of Social Capital on Users’ Future Contributions in an Online Health Community Xi Wang, University of Iowa, 810 W. Benton St B213, Iowa City, IA, 52246, United States, xi-wang-1@uiowa.edu Xi Wang, Central University of Finance and Economics, Beijing, China, xi-wang-1@uiowa.edu, Yuanyang Liu, Kang Zhao, Gautam Pant Online health communities are increasingly popular sources of information and support for people with health concerns. Encouraging users to make contributions on these platforms, however, is a challenge. Using an online health community for cancer survivors, this study examined if a user’s social capital could impact the user’s level of future contributions. The outcome reveals that different types of social capital may trigger a user’s future contributions in one or more ways. focus on detecting drug-related information from OHC data. 2 - Impact of Social Media on Copycat Suicides Jie Mein Goh, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada, jmgoh@sfu.ca, Srabana Dasgupta, Nilesh Saraf, Dianne Cyr 350F Social Media and Analytics – A Spectrum of Ideas from Diverse Industries Invited: Social Media Analytics Invited Session Chair: Fay Cobb Payton, North Carolina State University, Raleigh, NC, 27695, United States, fay_payton@ncsu.edu 1 - The Value of Social Media Data in Color Trends Forecasting and Inventory Decisions Youran Fu, University of Pennsylvania, 3730 Walnut Street, SC30

Room 532, Philadelphia, PA, 19104, United States, youranfu@wharton.upenn.edu, Marshall L.Fisher

We partnered with 3 leading apparel retailers to investigate how to use social media data to improve fashion color trend forecasting. We find that using fine- grained Twitter data and the Google search volume index to predict product-color sales three months out can significantly reduce forecast error compared to conventional methods.

77