2016 INFORMS Annual Meeting Program

SA43

INFORMS Nashville – 2016

SA40 207B-MCC Applications in Efficiency Analysis Invited: Data Envelopment Analysis Invited Session

The second part of the talk will explore differences in market dynamics over different time scales and price sensitivity to the shape of the order book. Efforts to find evidence for the concerns detailed in Part I will be explained. 3 - On A New Class Of Pairs-trading Strategies

Atul Deshpande, University of Wisconsin, Madison, WI, United States, adeshpande4@wisc.edu, B. Ross Barmish

The starting point for this paper is a new class of pairs-trading strategies which have rather weak requirements on the underlying stocks. We allow the user to define a rather arbitrary spread function to relate the two stock prices under consideration. Moreover, when this function, in conjunction with the price processes, satisfies a certain mean-reversion condition, we deem the stocks to be a tradeable pair. For such a case, we prove that our trading strategy results in positive expected growth in the account value. For a limited number of backtests using popular spread functions from literature, we see robust growth while avoiding huge drawdowns. 4 - Robust Empirical Optimization Is Almost The Same As Mean-variance Optimization Jun-ya Gotoh, Chuo University, jgoto@indsys.chuo-u.ac.jp We consider a robust optimization problem. Our main finding is that a large class of robust empirical optimization problems are essentially equivalent to an in- sample mean-variance. With this insight, we suggest a robust version of cross-validation which can be beneficial to high-dimensional-small-data situation such as portfolio optimization. SA42 207D-MCC Quantitative Methods in Finance Sponsored: Financial Services Sponsored Session Chair: Richard Sowers, University of Illinois, Champaign, Urbana, IL, 61801, United States, r-sowers@illinois.edu 1 - Measures Of Financial Network Complexity: A Topological Approach Mark Flood, Office of Financial Research, Mark.Flood@ofr.treasury.gov We present a general definition of complexity appropriate for financial counterparty networks, and derive several topologically based implementations. These range from simple and obvious metrics to others that are more mathematically subtle. It is important to tailor a complexity measure to the specific context in which it is used. This paper introduces measures of the complexity of search and netting in dealer markets. We define measures of line graph homology and collateral line graph homology that are sensitive to network interactions, such as collateral commingling and interdependent chains of obligations, that can be difficult or intractable to unwind.

Chair: Kankana Mukherjee, Babson University, 231 Forest Street, Babson Park, MA, 02457, United States, kmukherjee@babson.edu 1 - Malmquist Productivity Analysis With The Stoned Method Endre Bjørndal, Norwegian School of Economics, Endre.Bjorndal@nhh.no, Xiaomei Cheng, Mette Bjorndal We construct a Malmquist productivity index based on the Stochastic Non- parametric Envelopment of Data (StoNED) method, and we discuss how the distributional assumptions in the second StoNED stage affect estimates of productivity change and its decompositions. Our results are also valid for similar methods, such as COLS and MOLS, where a best-practice frontier is obtained from an average-practice frontier. Data on Norwegian electricity distribution companies is used to illustrate our discussion. 2 - Determinants Of Energy Efficiency In Indian Manufacturing Kankana Mukherjee, Babson University, kmukherjee@babson.edu This paper takes a production perspective and utilizes Data Envelopment Analysis to examine the major determinants of the traditional index of energy efficiency in the Indian manufacturing industry. The roles of technical efficiency of production, capacity utilization, and the adoption of energy efficient technology in determining energy efficiency is specifically addressed. 3 - Implicit Estimation For Stochastic Frontier Model Mohamed Nejib Ouertani, Sfax University, n_ouertani@yahoo.fr In this paper, a novel approach has been proposed to estimate stochastic frontier models; namely the Implicit approach. As we reckon, to our limited knowledge this method steel has not been applied to the stochastic frontier approach. For this purpose, the Markov Chain Monte Carlo (MCMC) techniques are used and implemented for both approaches subject of our study. As matter of fact, our new approach has been illustrated by means of a data pertaining to a sample of Saudian insurers. 4 - Inward Internationalization And Performance: An Application Of The Bootstrapped Truncated Regression Model Yi-Chun Liu, National Taiwan University, Taipei, Taiwan, d02724003@ntu.edu.tw, Yang Li, Homin Chen This study is the first to study the influence of the breadth of inward internationalization (BII) on the performance of internationalization of international tourists (ITHs). We employ the bootstrapped truncated regression model and takes into account the property of quasi-fixed inputs to analyze how BII influences the operational performance of Taiwan’s ITHs. The data set consists of 70 Taiwanese ITHs and 208 observations for the period 2011-2013. The empirical results show that the BII contributes positively to ITHs’ efficiency, benefits independent ITHs greater than it does for chain ones, and moderates the relationship between core revenue and operational performance. SA41 207C-MCC Control and Optimization in Finance Sponsored: Financial Services Sponsored Session Chair: James Primbs, Cal State Fullerton, Fullerton, CA, United States, jprimbs@fullerton.edu 1 - Order Book Dynamics And Price Sensitivity At Nasdaq: Part 1 James Primbs, California State University Fullerton, Fullerton, CA, 92831, United States, japrimbs@live.com Scott Condie, Sean Warnick The order book is a rich source of information detailing trading dynamics in equity markets. Recent concerns with algorithmic trading activities suggest that high frequency traders have unfair advantages compared with nominal traders. The first part of this talk will introduce the order book and explore the data set used in the analysis. Public concerns related to high frequency trading and Siren orders will also be discussed. 2 - Order Book Dynamics And Price Sensitivity At Nasdaq: Part 2 James Primbs, California State University Fullerton, 925 Berenice Dr, Brea, CA, 92821, United States, japrimbs@live.com Scott Condie, Sean Warnick The order book is a rich source of information detailing trading dynamics in equity markets. Recent concerns with algorithmic trading activities suggest that high frequency traders have unfair advantages compared with nominal traders.

SA43 208A-MCC Military Applications of Decision Analysis Sponsored: Decision Analysis Sponsored Session

Chair: Gregory S Parnell, University of Arkansas, University of Arkansas, Fayetteville, AR, 72701, United States, gparnell@uark.edu 1 - Stakeholder Requirements For Engineered Resilient Systems Decision Support Tools Christina Rinaudo, Operations Research Analyst, USACE Engineer Research and Development Center (ERDC), Vicksburg, MS, 39180, United States, christina.h.rinaudo@usace.army.mil, Randy Buchanan Engineered Resilient Systems (ERS) seeks to improve engineering and design development to generate adaptive, effective, and trusted systems in a physics model-based environment. The ERS program strives to incorporate decision analysis tools that enable customization of analytical tools. Through a collaborative workshop held with ERS stakeholders, researchers gained an understanding of multiple perspectives for decision analysis methods in a physics model-based environment. This presentation explores the stakeholder requirements identified by the workshop and proposes next steps to incorporate decision analysis into the ERS tradespace analytical tool suite.

31