Informs Annual Meeting 2017

MB21

INFORMS Houston – 2017

MB18

2 - Beyond Product Functionality Tian Chan, Emory University’s Goizueta Business School, 1300 Clifton Road, Atlanta, GA, 30322, United States, tian.chan@emory.edu Products combine function and form. We focus on form. We combine clustering techniques with experimental validation to identify styles (groupings of designs of similar form) among the over 350,000 US design patents granted from 1977 to 2010. Thus we compile a dataset of styles that can serve as an empirical platform to study the role of product form in NPD. Using the data, we find that (i) style turbulence (the unpredictability of changes in a style’s prevalence) follows a U- shaped relationship with respect to function turbulence and (ii) style turbulence increases over time. 3 - Corporate Predictions Markets: Evidence from Google, Ford and Firm X Bo Cowgill, Columbia Business School, 3022 Broadway, Uris Hall 702, New York, NY, 10583, United States, bo.cowgill@gmail.com We examine data from prediction markets run by Google, Ford Motor Company and an anonymous conglomerate (Firm X). We find these markets are relatively efficient and improve expert forecasts at all 3 firms by as much as a 25% reduction in mean squared error. The most notable inefficiency is an optimism bias in the markets at Google. The few inefficiencies that appear become smaller over time. Experienced traders and those with higher past performance trade against the identified inefficiencies. 4 - Value Creation and Capture in a World of Bottlenecks Douglas Hannah, University of Texas at Austin, Austin, TX, United States, dph@utexas.edu Ecosystems are settings in which firms depend on one another to provide the components that together comprise valuable products. Research has identified effective collaboration as central to firms’ ability to create value in ecosystems, but a gap exists with respect to when and for whom different types of collaboration are effective. I address this gap with a game theoretic model that compares two archetypal strategies, and thus contribute to research on ecosystems and collaborative strategy. 5 - Digital Platforms and Product Innovativeness: Towards an Understanding of Innovationat the Device Layer Hyunwoo Park, The Ohio State University, Columbus, OH, 30309, United States, park.2706@osu.edu The infusion of digital capabilities into physical products is an increasingly pervasive phenomenon. This prolific combination of “bits” and “atoms” necessitates a deeper understanding of the association between logical capability and physical machinery innovation. This chapter frames that digital platforms represent “areas” in technology design space that enable innovative recombination of technological components. It also proposes a novel and extensible index for quantifying innovativeness. 342C Predictive Analytics for Advancing Personalized and Evidence-based Medicine Invited: InvitedHealthcare Systems and Informatics Invited Session Chair: Eva Lee, Georgia Tech, Atlanta, GA, 30332-0205, United States, evakylee@isye.gatech.edu 1 - Machine Learning: Multi-site Evidence-based Best Practice Discovery Eva Lee, Georgia Tech, Industrial & Systems Engineering, Ctr for Operations Research in Medicine, Atlanta, GA, 30332-0205, United States, evakylee@isye.gatech.edu, Yuanbo Cody Wang This work is joint wih Care Coordination Institute. We invesigate best practice characteristics among 737 clinical sites that covers 2.7 million patients. Data interoperabiity is established via non-redundancy graph mapping onto standardized medical/lab/pharma terminologies. Next, unsupervised learning is performed to uncover treatment outcome patterns. Supervised learning via a classification model ((DAMIP) is then applied to uncover discriminatory characteristics that can predict the quality of treatment outcome. We will present results for diabetic patients. The patterns identified provide clinical practice guidelines for “rapid learning” across all clinical sites. MB21

340B Asymptotic Methods in Financial Mathematics Sponsored: Applied Probability Sponsored Session Chair: Jose Figueroa-Lopez, Washington University, St. Louis, MO, figueroa-lopez@wustl.edu 1 - Utility Maximization Problems in a Regime Switching Market Model Adriana Ocejo Monge, Assistant Professor of Statistics, University of North Carolina-Charlotte, Fret 335G, 9201 University City Blvd., Charlotte, NC, 28223, United States, amonge2@uncc.edu We present utility maximization problems where an agent builds her portfolio by investing in a bond and a risky asset whose price dynamics follow a diffusion process with regime-switching coefficients, modeled by a finite-state Markov chain. We consider the class of iso-elastic utility functions, which includes the power utility. We show how to explicitly compute the value function using Laplace transforms and present the optimal trading strategy. 2 - Optimal Thresholding of Truncated Realized Variations Jose E.Figueroa-Lopez, Washington University-St Louis, One Brookings Drive, St Louis, MO, 63130, United States, figueroa-lopez@wustl.edu, Cecilia Mancini We consider a univariate semimartingale model for (the logarithm of) an asset price, containing jumps having possibly infinite activity. We aim at optimally selecting the threshold of truncated realized variations by minimizing the conditional mean square error of the estimator. A parsimonious asymptotic characterization of the optimum is established, which turns out to be asymptotically proportional to the Levy’s modulus of continuity of the underlying Brownian motion. Moreover, minimizing the cMSE enables us to propose a novel implementation scheme for the optimal threshold sequence. Monte Carlo simulations illustrate the superior performance of the proposed method. 3 - Short Maturity Asian Options in Local Volatility Models Lingjiong Zhu, PhD, Florida State University, FL, United States, lz465@nyu.edu We present a rigorous study of the short maturity asymptotics for Asian options with continuous-time averaging, under the assumption that the underlying asset follows a local volatility model. The asymptotics for out-of-the-money and at-the- money cases are derived, and explicit formulas are given in the cases of Black-Scholes model, square-root model and CEV model. We present an analytical approximation for Asian options prices, and demonstrate good numerical agreement of the asymptotic results with the results of Monte Carlo simulations and benchmark test cases. This is based on the joint work with Dan Pirjol. 4 - Price Dynamics in a Limit Order Market under Time Dependent Order Flow Johnathan A. Chavez-Casillas, University of Calgary, Calgary, AB, Canada, jonathan.chavezcasil@ucalgary.ca In this talk we will, briefly, introduce Limit Order Books and explain the intricate dynamics between the order flow and the price process. We will discuss some efforts to describe the price dynamics and how they have been generalized to capture some empirical properties observed in the data. We will then introduce a model that considers a time-dependent order flow and, under some assumptions, characterize the price process within this model. We will then finish by describing how this model fit some particular data under the given assumptions.

MB20

342B TIMES Dissertation Award Sponsored: Technology, Innovation Management

& Entrepreneurship Sponsored Session Chair: Zhijian Cui, IE Business School, Calle de Maria de Molina, 12, 5th Floor, Madrid, 28006, Spain, Zhijian.cui@ie.edu 1 - Stochastic Models to Optimize Biomanufacturing Operations Tugce Martagan, Eindhoven University of Technology, Eindhoven, Netherlands, t.g.martagan@tue.nl We develop an inter-disciplinary framework to reduce biomanufacturing costs and lead times. The proposed framework consists of several Markov decision models that dynamically control and optimize protein purification operations. We partition the state space into decision zones, and propose a novel zone-based decision making approach to quantify the business risks and costs. We develop guidelines that are easy to implement in practice. Industry implementation indicates 20% reduction in protein purification costs and timelines.

171

Made with FlippingBook flipbook maker