2016 INFORMS Annual Meeting Program

409 / 561

INFORMS Nashville – 2016
409
3 - The Bravo Effect For Brownian Queues
Rob J Wang, Stanford University, ,
Peter W Glynn
In queueing theory, departure processes play a fundamental role. Indeed, for
single-station queues, they provide insights into system performance; for
feedforward networks, departures from one station constitute arrivals into the
next. Recently, there has been much interest in the asymptotic variability of
departure processes. The “BRAVO” (Balancing Reduces Asymptotic Variance of
Outputs) effect has been shown to manifest itself in many systems. This talk will
discuss BRAVO in the context of Brownian queues, which are tractable
approximations for various systems in heavy traffic. In particular, we will discuss
the timescales at which BRAVO appears, and offer explanations for its occurrence.
4 - Optimal Driver Routing In Crowdsourced Transportation Systems
Anton Braverman, Cornell University, Ithaca, NY, United States,
, J.G. Dai, Xin Liu, Lei Ying
We consider a queueing network that models the flow of drivers in a
crowdsourced transportation system such as Lyft or Uber. Each time a driver
drops off a passenger at their destination, a routing decision needs to be made.
Should the driver stay and wait for the next customer at their current location, or
should they drive empty to another part of town to try their luck there?
The way this decision is made greatly affects the supply of drivers across a city,
and can even cause extreme driver shortages in certain regions. We analyze the
fluid model corresponding to our network to develop a centralized routing policy
for drivers.
WB41
207C-MCC
Stochastic Control and Quantitative Finance
Sponsored: Financial Services
Sponsored Session
Chair: Xianhua Peng, Hong Kong University of Science & Technology,
Hong Kong,
1 - Leverage, Market Liquidity, And Systemic Risk
Nan Chen, Chinese University of Hong Kong,
We present a macroeconomic model with a financial intermediary sector subject
to a leverage constraint. The model allows us to examine the transition from
“normal” states to systemic risk states. An amplification effect through the
liquidity channel, both market and funding, can be characterized.
2 - Recursive Utility With Narrow Framing: Existence And Uniqueness
Xuedong He, Chinese University of Hong Kong,
We study the utility of an agent in a model of narrow framing with constant
elasticity of intertemporal substitution and relative risk aversion degree and with
infinite time horizon. In a finite-state Markovian setting, we prove that the utility
uniquely exists when the agent experiences nonnegative utility of gain and loss
incurred by holding risky assets and that the utility can be non-exist or non-
unique otherwise; in the latter case, we prove the existence and uniqueness with
further conditions. Moreover, we prove that the utility, when uniquely exists, can
be computed by a recursive algorithm with any starting point. Finally, we solve a
portfolio selection problem with narrow framing.
3 - Diversification Of Portfolio Tail Risk
Qi Wu, Chinese University of Hong Kong,
We develop explicit and accurate asymptotic expansions of the portfolio VaR and
portfolio Expected Shortfall (ES) for a large family of multivariate elliptical
distributions. We show that while the tail heaviness of joint asset distribution
dictates how much larger portfolio ES is comparing to VaR, it is the tail
dependence structure that determines the diversification benefits when sub-
portfolios are merged together for joint portfolio margining.
4 - An Empirical Likelihood Method Of Combining Stock And
Option Prices
Xianhua Peng, Hong Kong University of Science and Technology,
Steven Kou, Tony Sit, Zhiliang Ying
As discussed in the finance literature, option prices may contain information
about the dynamics of the underlying asset returns including the drift. In this
paper, we confirm this viewpoint by showing that the options information leads
to shorter confidence intervals for the parameters of the returns dynamics and
more efficient ways to reflect current market information. We propose an
empirical likelihood based method that can combine the stock return and the
associated derivative prices. Our empirical analysis of the S&P500 index and
options suggest that inclusion of option prices provides a more seasonable
estimates that can reflect the market conditions during the 2009 financial crisis.
WB42
207D-MCC
RM in Practice II
Sponsored: Revenue Management & Pricing
Sponsored Session
Chair: Xiaodong Yao, SAS Institute, Inc., SAS Campus Drive, Cary, NC,
27513, United States,
1 - Single-resource Capacity Control In The Presence Of
Cancellations, No-shows And Overbooking
Jason Chen, Principal Operations Research Specialist, SAS
Institute, Inc., Cary, NC, 27513, United States,
Single-resource capacity control problem is one of the most basic and well-studied
problems in quantity-based revenue management. Both exact methods and
efficient heuristic methods exist when cancellations and no-shows are ignored.
Dynamic-programming models have been proposed to solve the problem with
cancellation, no-show, and overbooking. But the DP model is only tractable for
small size problems. We present an efficient and highly scalable heuristic and
analyze its performance.
2 - Evolution Of Revenue Management Systems At Intercontinental
Hotels Group
Chanjoo Lee, InterContinental Hotels Group,
IHG’s PERFORM Price Optimization project started in 2005 to drive key strategic
priorities such as brand performance enhancement and excellent hotel returns.
The project was a large-scale enterprise implementation of price optimization in
the hospitality industry and provided a 2.7% increase in revenue as verified in
the IHG 2009 Annual Report. In this talk, we will discuss how the IHG Revenue
Management Systems including PERFORM Price Optimization evolved to increase
user acceptance from the hotels and drive revenue improvement over the years.
3 - Pricing And Revenue Management Of Function Space In Hotels
Xiaodong Yao, SAS Institute, Cary, NC, 27513, United States,
Altan Gulcu
Function space sales may provide significant contribution to the bottom line of
hotels, but implementing revenue management principles comes with additional
challenges. In this presentation, we review pricing and revenue management
techniques used for planning and management of function space in hotels.
WB43
208A-MCC
Decision Making with Incentives
Sponsored: Decision Analysis
Sponsored Session
Chair: Andrea Hupman Cadenbach, University of Missouri - St. Louis,
St Louis, MO, United States,
1 - Consider The Alternatives: New Ways To Finance
Early-stage Entrepreneurs
Samuel E Bodily, University of Virginia,
A startup business is ready for launch, yet the entrepreneur is unwilling to take
the considerable financial and potentially career-debilitating personal risk. We
compare game-changing alternatives to the usual equity model a backer may use
to finance the start-up: revenue contracts, derivative swaps, incentive gifts, and
insurance. We answer which best reduces the risk to the entrepreneur and
provides the best incentives, at a given cost to the backers, and without moral
hazard. Risk analysis models are used to compare certainty equivalents of these
financing alternatives.
2 - Nudging Vaccination With A No-fault Insurance
Emmanuel F Drabo, University of Southern California,
Los Angeles, CA, 90089-3333, United States, ,
Neeraj Sood, Joel W Hay, Jason N Doctor
Loss aversion in prospect theory and the K szegi-Rabin utility theory predicts that
insurance will be preferred to its expected value, hence implying that insuring
small risks of vaccine side effects can incentivize vaccine uptake. We test this
prediction through a discrete choice experiment with 1257 MTurk subjects
randomized into an insurance and a subsidy (expected value of insurance) group
to make choices among hypothetical vaccination scenarios. Vaccine uptake was
1.5 percentage points greater among non-female respondents in the insurance
arm relative to the subsidy arm. This suggests that a no-fault insurance against
vaccine side effects can be an effective vaccine incentive strategy.
WB43