2015 Informs Annual Meeting

WA07

INFORMS Philadelphia – 2015

2 - From Bermudan to American Liming Feng, Associate Professor, University of Illinois at Urbana- Champaign, 104 S Mathews Ave, Urbana, IL, 61801, United States of America, fenglm@illinois.edu The price of an American option could be approximated by that of a Bermudan option when the number of monitoring times is large enough. This talk presents results on convergence of the Bermudan option price and the early exercise boundary. Examples illustrate the effectiveness of pricing American contracts through the Bermudan approach. 3 - A Primal-dual Iterative Method for Stochastic Dynamic Programming Nan Chen, Prof, Chinese University of Hong Kong, 709A William Mong Engineering Building, Hong Kong, Hong Kong - PRC, nchen@se.cuhk.edu.hk, Xiang Ma We use the information relaxation technique to develop a value-and-policy iterative method to solve stochastic control problems. A sequence of upper and lower bound estimations on the true value functions and suboptimal policies are generated. We prove that the sequence converge to the optimal value and policy in a finite number of iterations. 4 - Modelling Time Dependence in Finance: A Time Change Approach Lingfei Li, Assistant Professor, The Chinese University of Hong Kong, 608 William M.W.Mong Engineering BLD, Shatin, Hong Kong - PRC, lfli@se.cuhk.edu.hk We propose a time change method named as additive subordination for constructing time-inhomogeneous Markov processes with tractability for financial modelling. As an application, we develop an analytically tractable model for pricing VIX options that is able to achieve excellent fit to the volatility surface. WA07 07-Room 307, Marriott Risk Management Approaches in Engineering Applications Cluster: Risk Management Invited Session Chair: Stan Uryasev, University of Florida, 303 Weil Hall, Gainesville, FL, United States of America, uryasev@ufl.edu 1 - Multidimensional Buffered Probability of Exceedance: Risk Averse Tail Probability Approximations Alexander Mafusalov, University of Florida, P.O. Box 116595, 303 Weil Hall, Gainesville, FL, 32611-6595, United States of America, mafusalov@ufl.edu Probability of exceedance (POE) is widely used but has major drawbacks. Buffered probability of exceedance (bPOE) is a conservative approximation of POE and eliminates some of its drawbacks. We suggest a new multidimensional generalization of bPOE (M-bPOE). M-bPOE is used to control exceedances for several random variables simultaneously. Calculation formulas and optimization formulations for M-bPOE are provided. Mathematical properties and possible applications for M-bPOE are studied. 2 - Soft Margin Support Vector Classification as Buffered Probability Minimization Matthew Norton, University of Florida, 2449 NW 93rd St, Gainesville, FL, 32606, United States of America, mdnorton@ufl.edu, Alexander Mafusalov We discuss a new concept called Buffered Probability of Exceedance (bPOE) and show that soft margin support vector classification is equivalent to bPOE minimization. This allows us to provide new insights into SVM’s. We provide a new SVM formulation, called the EC-SVM, based upon these insights. We also provide a general framework connecting soft margin support vector classification and superquantile concepts. 3 - Support Vector Machines Based on Convex Risk Functionals and General Norms Stan Uryasev, University of Florida, 303 Weil Hall, Gainesville, FL, United States of America, uryasev@ufl.edu, Jun-ya Gotoh We study unified formulations of support vector machines (SVMs) for binary classification, minimizing a combination of an arbitrary convex risk functional and an arbitrary norm-based regularizer. Especially, we investigate a condition under which nontrivial formulations can be obtained and what properties of risk functionals play roles in attaining interpretable dual formulations and/or tractable robust counterparts for robust optimization modelings.

4 - Energy Commodities: A Tool for Hedging and Investments Rita D’Ecclesia, Birkbeck, University of London, Department of Economics, Mathematics and Statistics, rita.decclesia@uniroma1.it Energy commodity prices in US and in Europe have shown increasing volatility, change in the relationship with other commodity prices and the traditional financial assets is also observed. Given the non stationarity of their returns we aim to measure the relationship existing between them and traditional financial assets using time varying volatility and correlation estimators. A portfolio diversification framework is also set using stochastic optimization modelling. WA09 09-Room 309, Marriott Innovation and Technology Management Contributed Session Chair: Zhongyuan Xu, Harbin Engineering University, 61th Building, 3046 Room, Harbin, China, xuruby0624@163.com 1 - Exploring Creative Collaboration in Maker Communities Yi-tzu Chang, Department of Public Relations and Advertising, The maker movement has drawn more and more attention from both management research and practice regarding information, technology, and innovation. Makers are creators who use digital technologies (e.g., 3D printing) to design and fabricate products collaboratively in DIY communities like FabLabs and makerspaces. This study interviews with 3D printing makers and develops a motivation-ability-opportunity framework to delineate creative collaboration in maker communities. 2 - How Much do Firms Earn in a Cooperative Innovation? From the Perspective of Relational Rent Si Zhang, University of Chinese Academy of Sciences, 1-1-23A Zhong Hai Fu Yuan, Zengguang Rd 37. Haidian District, Beijing, China, zhangsi@ucas.ac.cn, Jizhen Li, Na Li The paper first constructs an evaluation system that allows quantitative estimation and categorization of the relational rent generated by partners in collaborative innovation. Three types of rent are discovered, i.e. the individual, interactive and serendipitous rent. We then examine how the interdependence/relationship of focal paired firms affects the generation of rent.Valuable implications are provided to enterprises in the gaining and maximizing of benefits in innovative collaborations. 3 - Equity Split and Performance of High-technology Ventures: Whose Hands to Put the Company in? Yueheng Wang, Tsinghua Univeristy, Room 729B, Zijing Apartment No.14, Tsinghua University, Beijing, 100084, China, wangyh.08@gmail.com, Yanbo Wang, Jizhen Li The impact of equity split on firm performance is a crucial yet controversial topic regarding high-technology ventures in which shareholders play comparatively significant roles. With data of 626 Chinese high-technology ventures, we founded a curvilinear relationship between share concentration and firm performance, together with opposite moderating effects of board’s functional diversity and firm age, which reveal significant implications for shaping shareholder compositions within those firms. 4 - Innovation Networks and Innovation Performance: The Intermediary Role of Knowledge Management Capabi Zhongyuan Xu, Harbin Engineering University, 61th Building, 3046 Room, Harbin, China, xuruby0624@163.com This study extends the relationships between different innovation networks and innovation performance from a new perspective of knowledge management capability. Based on a survey to 239 Chinese knowledge intensive firms, this study empirically demonstrates how knowledge management capability plays an intermediary role from different innovation networks to innovation performance. 5 - Corporate Venture Capitalist Dependability and Their Inclusion in Venture Capital Syndicates Joseph Cabral, PhD Student, Rensselaer Polytechnic Institute, 124 Ferry Street, Apt. 213, Troy, NY, 12180, United States of America, cabraj2@rpi.edu Given that the venture capital industry involves great uncertainty, mutual dependency and staged investment, the reliability of partners is critical to the decision to co-invest. Using logic consistent with transaction cost economics Corporate Venture Capitalists are argued to be less predictable than traditional venture capitals and as a result are excluded from earlier rounds of investment where the greatest uncertainty and need for flexibility exists. Shih-Hsin University, No.1, Ln. 17, Sec. 1, Muzha Rd., Wenshan, Taipei, Taiwan - ROC, silverzal315@gmail.com, Cheng-chieh Hsiao

376