Informs 2015

69 / 552

INFORMS Philadelphia – 2015
67
SB04
04-Room 304, Marriott
JFIG Paper Competition II
Sponsor: Junior Faculty Interest Group
Sponsored Session
Chair: Asoo Vakharia, Professor, University of Florida, Department of
ISOM, Gainesville, FL, 32611, United States of America,
Chair: Arda Yenipazarli, Assistant Professor of Operations Management,
Georgia Southern University, COBA 2224, Statesboro, GA, 30460,
United States of America,
The 2015 JFIG paper competition features paper submissions from a diverse array
of talented junior faculty members. The prize committee evaluated submissions
based on the importance of the topic, appropriateness of the approach, and
significance of the contribution. After careful review, the prize committee selected
a group of finalists to present their research in one of the two JFIG sessions. For
information on the finalists and their papers, please refer to the online program.
SB05
05-Room 305, Marriott
Bridging Business and Analytics
Cluster: Social Media Analytics
Invited Session
Chair: Mingfei Li, Associate Professor, Bentley University, 175 Forest
Street, Waltham, United States of America,
1 - Implement Business Analytics in Education: An Exploration in
Business Education
Mingfei Li, Associate Professor, Bentley University, 175 Forest
Street, Waltham, MA, United States of America,
Do we have a formula for Business Analytics in education? How do we fill the
gap between business and analytics in school? Ideas of combining Business with
Analytics in education will be explored. Using Bentley’s business analytics
programs as an example to show works from students in this field.
2 - The Analytics of Kickstarter
David Oury, Lecturer, Bentley University, 175 Forest Street,
Morison room 325, Waltham, MA, 02451,
United States of America,
Our project investigates those factors that influence the success or failure of
crowdfunding projects through analytics on the Kickstarter data set. We compare
results, models and capabilities of the following software packages: Alteryx, IBM
SPSS Modeler, KNIME, R, RapidMiner, Salford Systems, SAS Enterprise Miner
and Semcasting.
3 - Analytics, Interdisciplinary Collaboration, and the Mapping of
Brand DNA
Joseph Dery, PhD Student, Sr. Data Scientist, Bentley University,
EMC Corporation, 175 Forest Street, Waltham, MA, 02451,
United States of America,
Do successful analytics projects utilize interdisciplinary collaboration? Leveraging
the brand-exclusivity robbing phenomenon of brand genericide, this deceivingly
marketing-dependent problem is viewed through a combined lens of brand
management, legal studies, linguistics and analytics. The result: newly mapped
“Brand DNA” with the ability to measure brand-mutations through longitudinal
text mining.
4 - An Exploration of Power Structures Utilizing Network Analysis
Kevin Mentzer, PhD Student, Bentley University, 175 Forest
Street, Waltham, MA, 02451, United States of America,
, Dominique Haughton
This work combines Bayesian analysis with social network analysis to highlight
significant change in networks over time. Applications exploring power in
interlocked corporate boards as well as power in state government are explored.
Findings show capabilities to identify and isolate key data as well as show shifting
power dynamics.
SB06
06-Room 306, Marriott
Stochastic Systems in Finance
Sponsor: Financial Services
Sponsored Session
Chair: Alexandra Chronopoulou, Assistant Professor, University of
Illinois at Urbana-Champaign, 117 Transportation Building, MC-238,
104 South Mathews Avenue, Urbana, IL, 61801,
United States of America,
1 - Efficient Risk Analysis for Loan Pools
Justin Sirignano, Stanford University, Huang Engineering Center,
Stanford, Ca, 93404, United States of America,
Kay Giesecke
Financial institutions and investors are often exposed to default risk from large
numbers of loans. Due to the size of loan pools, brute-force simulation is
computationally expensive. We prove weak convergence results in order to
construct an efficient Monte Carlo approximation. We test our approximation on
a data set of over 25 million actual mortgages. Computational cost is often several
orders of magnitude less than brute-force simulation of the actual pool with a
similar level of accuracy.
2 - Optimally Thresholded Realized Power Variations for Stochastic
Volatility Models with Jumps
Jose Figueroa-lopez, Purdue University,
Thresholded Power Variations are popular nonparametric estimators for
continuous-time processes with jumps. An optimal threshold selection approach
is put forward in the presence of a stochastic volatility risk component. To this
end, we further develop current kernel based estimators for the spot volatility,
which in turn yield new optimal bandwidth selection procedures for stochastic
volatility models.
3 - Principal Component Analysis of High Frequency Data
Dacheng Xiu, University of Chicago, Booth Business School,
Chicago, IL, Yacine Ait-sahalia
We develop a methodology to conduct principal component analysis at high
frequency. The procedure involves estimation of realized eigenvalues, realized
eigenvectors, and realized principal components and we provide the asymptotic
distribution of these estimators.
4 - Indifference Pricing for Contingent Claims:
Large Deviations Effects
Konstantinos Spiliopoulos, Assistant Professor, Boston University,
Department of Mathematics and Statistics, 111 Cummington
Mall, Boston, MA, 02215, United States of America,
We study utility indifference prices and optimal positions for a non-traded
contingent claim in an incomplete market with vanishing hedging errors, making
connections with large deviations. Consider a sequence of semi-complete markets
where for each n the claim hn = Dn+Yn, Dn is replicable and Yn is unheadgable.
In this setting, we show the prices typically are not the unique arbitrage free price
in the limiting market and that optimal purchase quantities occur at the large
deviations scaling.
SB07
07-Room 307, Marriott
Networks and Contagion Risk
Cluster: Risk Management
Invited Session
Chair: Agostino Capponi, Columbia, Mudd 313, New York, NY, 10027,
United States of America,
1 - Interfirm Relationships and Asset Prices
Carlos Ramirez, Finance PhD Candidate, Carnegie Mellon
University, 5000 Forbes Ave, Pittsburgh, 15213,
United States of America,
I study asset pricing in large economies where persistent interfirm relationships
generate interdependencies among firms’ cash-flows. In the calibrated model:
well connected firms command higher risk premium than less connected firms;
firm-level return volatilities follow a factor structure; and momentum trading
strategies are profitable.
SB07