2015 Informs Annual Meeting

WD06

INFORMS Philadelphia – 2015

WD06 06-Room 306, Marriott Portfolio Analysis I Contributed Session

2 - Data Science Approach for Dealership Performance Analysis Haidar Almohri, Wayne State University, Detroit, MI, 48243, United States of America, almohri@wayne.edu, Mark Colosimo, Ratna Babu Chinnam Due to the dynamics and complexity of the automotive market, managing automotive dealership performance has always been a challenge for business owners. We present a novel, data science approach for managing, assessing, and enhancing dealership performance using data from dealerships across the United States by first clustering the dealers. Results suggest that this new approach surpasses the traditional approach employed by the industry. 3 - Overview of Supply Chain Risk Management and the Current Issues of Closed-loop Supply Chain in China Qiang Qiang, Pennsylvania State University, 30 E. Swedesford Rd., Malvern, PA, 19335, United States of America, qzq10@psu.edu The path toward sustainability to demonstrate environmental and social responsibility has led to an increasing attention to the lifecycle of a product with a focus on value-added recovery activities. However, besides the risks encountered in traditional supply chains (or, sometimes referred to as the forward supply chain), the CLSC bears additional layer of risks if the products are not properly recycled. In this talk, we will provide an overview of both supply chain and CLSC risks in China. 4 - Supply Chain Risk James Ang, Research Advisor, The Logistics Institute Asia Pacific, National University of Singapore, #04-01 21 Heng Mui Keng Terrace, Singapore, 119613, Singapore, bizangsk@nus.edu.sg Supply chain risk analysis involves risk identification assessment. For the former, the failure mode effect analysis is improved using fuzzy theory and grey system theory. For the latter, simulation models of supply chains are built to quantitatively assess impacts of risks in terms of production rate, inventory level, and lead-time. WD05 05-Room 305, Marriott The Social Impacts of Social Media Analysis Cluster: Social Media Analytics Invited Session Chair: Les Servi, The MITRE Corporation, 202 Burlington Road, Bedford, MA, United States of America, lservi@mitre.org 1 - Modeling Human Behavior in the Context of Social Media During Extreme Events This research examines how social science theory, Theory of Planned Behavior, can explain human behavior in response to extreme events on social media. Validation of this theory enables emergency response officials to create strategies that facilitate public response to extreme events such as diffusion of critical actionable information, providing confirmations, and taking the prescribed action. Effective public response can save lives and reduce property damage. 2 - Social Media Whispers during IPO Quiet Period Julie Zhang, Assistant Professor, University of Massachusetts We investigate social media news for companies during their quiet periods for IPO. In this study, we examine whether the stock price of a company on its IPO day is associated with the volume and valence of its social media coverage during quiet period. We demonstrate the relationship between a company’s stock price PARK volatility in the first week of IPO and its social media coverage in the first week and quiet period. 3 - Simulating Twitter Ego Networks Inbal Yahav, Bar Ilan University, Bar Ilan University, Ramat Gan, 5290002, Israel, yoavac@gmail.com,Yoav Achiam The network on the 2-level (followers, and their followers) Twitter social ego- network is studied. A simulation based on A. Barabasi’s model is generated. The collected data contains the followers’ names and a count of their followers. We found that followers’ distribution follows a combination of power law and exponential distributions. The distribution of followers per companies (having more than 1000 followers) is steady over time, with a constant rate of new followers and departing followers Yulia Tyshchuk, Researcher, USMA, 5 Winding Brook Dr. Apt. 2J, Guilderland, NY, 12084, United States of America, yulia.tyshchuk@gmail.com Lowell, One University Ave., Lowell, MA, 01854, United States of America, juheng_zhang@uml.edu

Chair: Umit Saglam, Assistant Professor, East Tennessee State University, Department of Management and Marketing, College of Business and Technology, Johnson City, TN, 37614, United States of America, saglam@etsu.edu 1 - Understanding Behaviors of Robust Portfolios Woo Chang Kim, Associate Professor, KAIST, 291 Daehak-Ro, Yuseong-Gu, Daejeon, Korea, Republic of, wkim@kaist.ac.kr Robust portfolio optimization has been developed to resolve the high sensitivity to inputs of the Markowitz mean–variance model. Although much effort has been put into forming robust portfolios, there have not been many attempts to analyze the characteristics of portfolios formed from robust optimization. In this presentation, we discuss the recent findings on the qualitative characteristics of the robust portfolios such as higher moment controllability, factor tilting behaviors, and robustness. 2 - Computing Near-optimal Value-at-risk Portfolios using Integer Programming Techniques Onur Babat, PhD Student, Lehigh University, 217 W. Packer Ave, VaR is a non-convex risk measure. It is well-known that the VaR portfolio problem can be formulated as an integer program (IP), which can be difficult to solve with current IP solvers for large-scale instances of the problem. To tackle this drawback, we present an algorithm to compute near-optimal VaR portfolios that takes advantage this IP formulation and provides a guarantee of the near- optimality of the solution. Numerical results will be presented. 3 - Portfolio Optimization with Probabilistic Ratio Constraints Ran Ji, PhD Candidate, The George Washington University, 2201 G St, NW, Funger 415H, Washington, DC, 20052, United States of America, jiran@gwmail.gwu.edu, Miguel Lejeune We propose a class of stochastic portfolio optimization models with probabilistic ratio constraints. The proposed probabilistic reward-risk ratio measures regard the asset returns/losses as random variables to mitigate the estimation risk due to mean return vectors. Each model includes a chance constraint with random technology matrix. We expand a combinatorial modeling framework to represent the feasible set of the chance constraint as a set of mixed-integer linear inequalities. 4 - A Marginal Conditional Stochastic Dominance Based Model of Enhanced Index Tracking Qian Li, Professor, Xi’an Jiaotong University, No.74 West Yanta Road, Xi’an, China, lqian@mail.xjtu.edu.cn, Liang Bao Stochastic dominance and mean-variance are two criteria used in optimal portfolio selection. In this paper, we present an enhanced model for index tracking with Marginal Conditional Stochastic Dominance (MCSD) rules. The model is still in the optimization framework but combined with two levels of MCSD criteria. By adopting an immunity based multiple objective optimization algorithm, the solutions for the model are developed. The model is then applied to 8 major markets. 5 - Revealed Preferences for Portfolio Selection – Does Skewness Matter? Umit Saglam, Assistant Professor, East Tennessee State University, Department of Management and Marketing, College of Business and Technology, Johnson City, TN, 37614, United States of America, saglam@etsu.edu, Merrill Liechty In this study, we consider two competing descriptions of portfolio selection, the traditional mean variance efficient portfolio versus a generalization allowing for decision makers to consider skewness in their asset allocation. Our numerical experiments are conducted on portfolio drawn from 30 different stocks from the Dow Jones. Numerical results show that investors’ preferences are better explained when skewness is taken into account. Bethlehem, PA, 18015, United States of America, onur.babat@lehigh.edu, Juan Vera, Luis Zuluaga

455