Informs Annual Meeting 2017

MA55

INFORMS Houston – 2017

MA55

We study recharging infrastructure design problem for the urban areas. Considering the investor and consumer viewpoints along with the inherent uncertainty about the recharging demand, we approach to this problem from a broad perspective to obtain significant managerial insights that can help decision makers to shape long term strategies to take advantage of the new opportunities offered by the emerging electric vehicle technologies. We present a branch and cut algorithm to find optimal infrastructure design that minimizes the total investment cost while guaranteeing a target quality of service level that is crucial Burcu B. Keskin, University of Alabama, 4138 Meretta Lane, Tuscaloosa, AL, 35406, United States, bkeskin@cba.ua.edu, Mesut Yavuz In this work, we study highway patrol by state troopers, which differs from police patrol in regards to (i) geographical coverage, (ii) miles traveled by a vehicle, and (iii) the higher probability of involvement in high-speed, long-distance chases. Specifically, for a given mixed patrol fleet consisting of traditional (gasoline or diesel) vehicles and HEVs, we investigate optimal patrol routes to visit time- critical hot spots. Our overall goal is to maximize the visibility of the state troopers while minimizing the costs associated with utilization of troopers. We analyze mathematical properties of the problem to offer analytic solutions or transformations to polynomially-solvable problems. 3 - Random Utility Model for Electric Vehicle Movement with Respect to Multiple Charging Stops Yudai Honma, The University of Tokyo, Ce405, Komaba 4-6-1, Meguro-ku,, Tokyo, 153-8505, Japan, yudai@iis.u-tokyo.ac.jp, Shigeki Toriumi This research focuses on the EV-support infrastructure of charging facilities in terms of EV route-decision behavior and multiple charging stops. When considering the movements of long-distance trips, it is very important to incorporate the route of EV movements because there is not only a simple and shortest path from origin to destination, but also several alternative paths. Furthermore, it is also essential to consider an EV’s multiple stops at EV- recharging facilities. An elegant mathematical model to analyze the EV-support infrastructure with respect to multiple routes and multiple-charging stops will be discussed. 4 - A Heuristic Method to Solve Large-scale Flow-refueling Location Model Ibrahim Capar, Bowling Green State University, Department of ASOR, 355 Business Administration Building, Bowling Green, OH, 43403, United States, icapar@bgsu.edu, Ismail Capar In this research, we propose a new heuristic model for flow-refueling location model for alternative fuel vehicles. One of the challenges many organizations face is using existing exact and heuristic algorithms to solve large scale real-world problems. We present a heuristics approach to solve these larger scale problem consisting of real-world data set obtained from an organization located in Europe. We discuss the heuristic in detail and share the results from numerical experiments. 362D Joint Session SIM/ OPT under Uncertainty: Sample Average Approximation (SAA): Applications and Methodology Sponsored: Simulation Sponsored Session Chair: Dashi Singham, Naval Postgraduate School, Naval Postgraduate School, Monterey, CA, 93943, United States, dsingham@nps.edu 1 - Sample Average Approximation for Continuous Principal Agent Problems Dashi Singham, Naval Postgraduate School, 1411 Cunningham Road, Operations Research Department, Monterey, CA, 93943, United States, dsingham@nps.edu, Wenbo (Selina) Cai We study a sample average approximation for estimating the solution to a continuous-demand principal-agent problem. This approach uses discrete samples from the demand distribution to solve a deterministic demand version of the problem. A special case of the continuous-demand problem where the solution is known motivates the problem. The broader issue is the convergence of functional solutions to SAA problems. 2 for the transportation and logistic use of electric vehicles. 2 - Patrol Routing with Hybrid Electric Vehicles MA57

362B Financial Econometrics Sponsored: Financial Services Sponsored Session Chair: Markus Pelger, Stanford University, Stanford, CA, 94305, United States, mpelger@stanford.edu 1 - Taming the Factor Zoo Dacheng Xiu, PhD, Chicago Booth, Chicago, IL, United States, Dacheng.Xiu@chicagobooth.edu The asset pricing literature has produced hundreds of potential risk factors. Organizing this ``zoo of factors” and distinguishing between useful, useless, and redundant factors require econometric techniques that can deal with the curse of dimensionality. We propose a model-selection method to systematically evaluate the contribution to asset pricing of any new factor, above and beyond what a high-dimensional set of existing factors explains. Our procedure selects the best parsimonious model out of the large set of existing factors, and uses it as the control in making statistical inference about the contribution of new factors and estimating their price of risk. 2 - Efficient Parameter Estimation for Multivariate Jump-diffusions Gustavo Schwenkler, Boston University, 595 Commonwealth Ave, Boston, MA, 02215, United States, gas@bu.edu, François Guay This paper develops an unbiased Monte Carlo estimator of the transition density of a multivariate jump-diffusion process. The coefficient functions are allowed to be state-dependent and non-affine. Our estimator facilitates the parametric estimation of multivariate jump-diffusion models based on discretely observed data. The parameter estimators we propose have the same asymptotic behavior as maximum likelihood estimators as the number of data points grows, but the observation frequency of the data is kept fixed. Our density and parameter estimators are found to be highly accurate and computationally efficient in a numerical study. 3 - Functional Autoregression for Sparsely Sampled Data Daniel R.Kowal, Cornell University, Ithaca, NY, United States, drk92@cornell.edu Daniel R.Kowal, Rice University, Houston, TX, United States, drk92@cornell.edu, David S. Matteson, David Ruppert We develop a hierarchical Gaussian process model for forecasting and inference of functional time series data. The latent process is dynamically modeled as a functional autoregression with Gaussian process innovations. For the dynamic innovation process, we propose a fully nonparametric dynamic functional factor model, with broader applicability and improved computational efficiency over standard Gaussian process models. Extensive simulations demonstrate substantial improvements in forecasting performance and recovery of the autoregressive surface over competing methods. We apply the proposed methods to forecast

nominal and real yield curves using daily U.S. data. 4 - Asset Pricing Testing for Many Assets

Markus Pelger, Stanford University, 312 Huang Engineering Center, 475 Via Ortega, Stanford, CA, 94305, United States, mpelger@stanford.edu, Luyang Chen

We derive the asymptotic inferential theory for cross-sectional asset pricing testing when both the number of time observations T and number of assets N are large. After proper normalization the time-series and cross-sectional test statistics are asymptotically normal distributed in contrast to the chi-squared distribution for fixed N. Conventional time-series and cross-sectional tests can lead to contradicting conclusions for large N. Even for moderately sized cross-sections with N = 75 conventional test statistics are strongly biased, while our test statistics have good power and size properties.

MA56

362C Fueling Station Location and Routing Decisions for Alternative Fuel Vehicles Sponsored: Location Analysis Sponsored Session Chair: Ismail Capar, Texas A&M University, College Station, TX, 77843-3367, United States, capar@tamu.edu Co-Chair: Ibrahim Capar, Bowling Green State University, Bowling Green, OH, 43403, United States, icapar@bgsu.edu 1 - The Recharging Infrastructure Design Problem with Capacitated Recharge Stations and Stochastic Demands Baris Yildiz, Koc University, Istanbul, 34450, Turkey, byildiz@ku.edu.tr, Evren Olcaytu

152