Info-Metrics Institute
College of Arts and Sciences, American University
Fall 2014 Newsletter
Director’s Update
Contents
Director’s Update............................1
Notes from the Advisory Board.....2
Institute World Impact.......................3
Special Session on Info-Metrics....3
Entropic Inference...........................3
Research Updates......................4-10
Seminar of the Year..........................5
XacBank Mongolia..........................6
Visiting Fellows.............................10
Forthcoming Tutorials.................10
Thank You To Sponsors.................10
Affiliate Collaborations................10
Affiliate Profiles.............................11
Grad Student Words......................12
Institute Events..............................12
Support the Institute......................12
Info-Metrics Institute
Department of Economics
Kreeger Hall
4400 Massachusetts Ave, NW
Washington, D.C. 20016-8029
info-metrics@american.edu
www.american.edu/info-metrics
The Info-Metrics Institute is celebrating its fifth year. The original vision was simple: establish a
small institute to bring together
researchers from across the scientific spectrum to study together all aspects of info-metrics, the
science, practice and art of processing information. Info-metrics provides a common thread
among all disciplines and decision makers since they all proPanel table discussion (L-R: Eric Renault, Michael Stutzer, Wojciech Szpanzowkski,
cess information constantly. It
Raphael Levine) at the Information, Instability and Fragility in Networks: Methods and
is universal. But to really underApplications workshop in November of 2013 at Hotel Boulderado in Boulder, CO
stand and study info-metrics we
must develop a common language, learn from each other and study together the foundations of information and information
processing at all levels: from the more philosophical to the more quantified; from theory to practice; and from one discipline
to another. Through the Institute, collaborations across disciplines and researchers are fostered via workshops, conferences,
seminars, tutorials, visiting fellowships and inclusive research projects.
Five years later, it is quite satisfactory to observe that the Institute’s achievements and reputation have exceeded our wildest
dreams. The Institute has hosted ten workshops and conferences (in all areas of info-metrics and in different locations), many
seminars and tutorials, a large number of visiting fellows (at all levels), as well as twelve summer graduate student fellowships
(from different universities). But even more impressive than the sheer count of the Institute’s activities, is the success it has
had in disseminating the science of info-metrics. Researchers across disciplines and continents are turning their interest to
the study and practice of info-metrics. Students across disciplines and at all levels of study are being exposed to info-metrics,
from its theoretical foundations to its application. The Institute is helping to transform info-metrics into a true sub-discipline.
Building on our momentum and infrastructure, the Advisory Board and I plan to move to the next phase of pushing the
research frontiers of info-metrics and enhance our role as a leading interdisciplinary institute. We will do it in a number of
ways. First, we will continue our efforts to reach (and engage with) young and established researchers from across the sciences,
especially from the natural, medical and engineering sciences. Second, we will expand our efforts to initiate and cooperate on
activities outside the Washington DC area, including different continents. For example, our spring conference will take place
in Oxford, UK. Third, we will work to improve our tutorials and consider providing more tutorials at different locations and
in a more interdisciplinary way. Fourth, we will start to discuss a potential interdisciplinary graduate level program. Though
it seems that there is much demand for such a program, the main issue for us is whether we want to expand our activities in
that direction, and if so, in what way and where.
Please check the Institute’s website frequently for information about our activities.
Ideas for new initiatives are always welcome!
Upcoming Institute Events
Fall 2014 Conference:
As always, we thank the Office of the Comptroller of the Currency (OCC) for its Recent Innovations in Info-Metrics
generous, continued support. The recent long-term grant from the OCC allows the October 31 & November 1, 2014
Institute to continue strengthening its role as a leading interdisciplinary institute
American University
helping to shape future research, supporting and tutoring graduate students, and
introducing new generations of researchers across disciplines to the theory and
Spring 2015 Workshop:
practice of info-metrics. We also thank our affiliate Mike Stutzer and the University
Philosophy of Information and
of Colorado at Boulder for organizing (and co-sponsoring) our November conference
Information Processing
at Boulder, CO. We also thank XacBank, Mongolia for co-sponsoring this conference.
March 27, 2015
I would also like to thank you all for your continued support and interest in the Info- Pembroke College, Oxford, UK
Metrics Institute. Our efforts to establish a common language to link disciplines in
solving similar info-metric problems are showing nice results. Luckily, there is still
For more information,
much to do and I look forward to working with you all in the coming years.
please visit our website.
-Amos Golan, Director, Info-Metrics Institute
2
Info-Metrics Institute Fall 2014 Newsletter
Notes from the Advisory Board
The first time I participated in an InfoMetrics Institute workshop I seemed to
hear John Denver singing, “Coming home
to a place he had never been before.” When
I first heard of the Info-Metrics Institute
I thought of it as spun-off from the
Department of Economics at American
University, focusing on information
and economics. I had not realized that
Jon Michael Dunn, member while Amos Golan was a Professor of
Economics, he had, as they say in Monty
Python, created “something completely different.” Info-metrics is the
quantitative measure of information, of whatever kind – not just economic.
Throughout my career I had been working on information-based logics:
relevance logic, intuitionistic logic, quantum logic, etc. But I never quite
saw “the big picture” until around 1999 when I became the founding Dean
of the School of Informatics at Indiana University and I began to understand
INFORMATION writ large. I think I am thus in a special position to
appreciate the creativity and plain hard work that Amos has expended.
My research, like all of the Info-Metrics affiliates, has been positively
affected because of the input received from the workshops I have attended,
and because of my being a visiting fellow at the Institute last October I have
been working with Amos Golan on the value of information. Although I am
now retired from Indiana University I am still very active in my research,
and very proud to be on the Info-Metrics Institute Advisory Board.
info-metrics [in-fo-MET-riks] noun: the science and
practice of inference and quantitative information processing.
Changing a paradigm is often a long and
ambitious process. It takes intellectual
leadership, but also the actions of a
community of scholars to diffuse the
concepts, apply them to a range of
problems, and teach others. The InfoMetrics Institute offers a case study in
pushing a new paradigm forward. In this
case, a key figure in the development of
info-metrics, Professor Amos Golan, is
Robert Lerman, Chair
also the founder, fundraiser, organizer,
and builder of an institutional base for
expanding the reach of info-metrics. Under his leadership, the Institute
has helped make info-metrics a well-known discipline that spans a range
of scientific fields, including biology, economics, philosophy and physics. The Info-Metrics Institute has brought together an extraordinary group
of internationally recognized scholars for conferences and for research
fellowships, taught scores of students, provided summer fellowships for
in-depth learning by four graduate students, and has stimulated research
that is pushing the frontiers of statistical analysis. In only five years, the
Institute has played a critical role in building the community of scholars
that is changing a paradigm. These developments could not have taken
place without the resources provided by our sponsors, most importantly
the Office of the Comptroller of the Currency.
The Info-Metrics Institute is continuing to expand its network of active
scholars and its reach to the international science community. The March
2015 Institute conference in Oxford, England, will widen the exposure of
European researchers to the concepts and applications of info-metrics. With
the continuing strength of the Institute and the compelling attractiveness of the
info-metrics outlook, the Institute community can feel proud to be contributing
to a paradigm shift in the way scientists approach data and information. Esfandiar Maasoumi, member
I have had the privilege of being involved
and associated with the activities of the InfoMetrics Institute since its inception. It is fair
to say that all of its ideals, expressed and hoped
for, have been realized in multiple ways. These
include exceptionally well-conceived and
organized conferences on multidisciplinary
themes, enveloped by information theory
conceptions and methods.
The Institute and its activities have provided focus and a reliable forum in which
many scholars can present and test their ideas and demonstrate their value and
potential across multiple disciplines. The Institute and its scholarly programs
have reflected well the universality of information, its promise and challenges.
I have been very fortunate to benefit from the ideas and interactions with
other members and participants. This has helped maintain a consistent
and coherent element in my own research work, ranging from forecasting
and model evaluation, to a whole host of measurements and applications in
income inequality, poverty and mobility. All of these areas require concepts
and metrics for assessment of distributed outcomes, and information theory
is unchallenged as the primary source for management of these challenges.
The latest example of the ways in which our work at the Institute has
informed very topical, current debates, is the question of the “gender gap”.
The very definition of it demands better conceptualization than differences
in means and medians, calling for entropic measures that represent the
entire distribution of wages or wealth. The master econometricians
who defined the very subject matter were quite informed and aware
of this potential, and the ability of information theoretic methods for
characterizing and assessing “allocation” and “aggregation” questions.
These have reoccupied their rightful central place in economic and
political debates of our time, and the Institute can look forward to playing
an effective and constructive role in this crucial development.
Some of the work that is presented in my recent papers with fellow
member, Jeff Racine, was conducted while we were both attending
conferences at the Institute or as visiting scholars. This is partly
reflected in the paper, ”A Solution to an Aggregation Issue”, in
which nonparametric estimation is employed to derive estimates of
fundamental multidimensional objects that measure the illusive concept
of wellbeing. The gender gap work with Le Wang is partly presented in
the paper, “The Gender Gap: Analysis and Measurement.”
It is commonly understood by leading members of the Institute that
this enterprise, like all others like it, depends on dedication and tireless
work of a leader. Amos Golan has been that leader from the start and
it is easy to underestimate the effort and work that he has expended to
ensure the success and vitality of the Institute.
www.american.edu/info-metrics
Info-Metrics World Impact
Among Institute Affiliates
Rossella Bernandini Papalia
The Info-Metrics Institute promotes the
development of theoretical tools and
applications in information theory. The
workshops, conferences, and courses
organized by the Institute provide an
excellent platform for disseminating
the results of the research and encouraging discussion of ideas in
academic and research institutions and in society at large. I gained
knowledge from common services offered as resources on the
Institute’s website such as special journal issues, working papers,
and software. I found very fruitful attending the workshops
and conferences and visiting the Institute in recent years.
My current research on small area estimation and spatial disaggregation
approaches became a strategic field in applying IT methods in the field
of ecological inference that also account for data autocorrelation and
heterogeneity. The main challenges for this methodological research
area are in the integrated use of data sets coming from various sources
with heterogeneous quality. Administrative and other register data
are increasingly used in integrated statistical systems and further
methodological development is needed in this field.
Statistical agencies and data producers will be required to try out new
estimation approaches and to provide theoretical developments, which
enable a coherent view on the solutions for particular questions connected
to the production of data by addressing more complex sampling situations
such as spatial and temporal sampling, geo-referencing and sampling of
rare events and dynamic populations, as well as non-random samples.
This research activity was also developed within the BLUE-ETS work
programme (http://www.blue-ets.istat.it/), financed by the European
Commission, under the 7th Framework Programme.
My recent research within the IT framework has aimed at developing
(i) multidimensional economic performance indexes with a microlevel foundation and (ii) general composite indicators that are based
on spatial structural equation models.
Michael Stutzer
Nearly 20 years ago, I published an
entropic diagnostic for quickly identifying
flaws in asset pricing models developed
by financial economists. The paper was
published in the Journal of Econometrics
(“A Bayesian Approach to Diagnosis of
Asset Pricing Models” 1995, pp. 367-397)
and therefore, the method was not adopted by business school finance
faculty who mainly study articles published in a few top finance journals. This year, a closely related article appeared in one of those journals,
creating a revival in the subject among younger researchers. When I
contact them, they have been surprised to find out about the work that
others and I in the Info-Metrics Institute have done and are still doing. I hope this will create a new cohort of talented researchers that will
contribute to the Info-Metric Institute’s mission. 3
Info-Metrics at the Eastern
Economic Association Conference
On March 9, at the 2014 Eastern Economic Association Conference
in Boston, four American University Economics PhD students
(Paul Corral, Mungunsuvd Terbish, Ermengarde Jabir, and Daniel
Kuehn) presented their fall 2013 seminar work on info-metrics.
Despite daylight savings time and an early morning time slot, the
session was well attended. Paul Corral and Mungunsuvd Terbish began the session with a general
discussion of generalized maximum entropy (GME) and presented a
new Stata command written by them for a GME discrete choice
model. Their paper is currently under review at The Stata Journal.
Ermengarde Jabir presented her work on household asset allocation
using GME estimation methods. Daniel Kuehn presented his work on
a GME version of propensity score matching (PSM), a common quasiexperimental technique in the evaluation literature. Currently, Daniel
is comparing GME PSM to traditional approaches using simulations,
and he hopes to eventually apply it to actual program data.
Recently, the same team has been accepted to present as a panel at
the 2014 Southern Economic Association Conference in Atlanta
this November. This year, the team will be joined by Heath
Henderson, an American University Economics Department
graduate, who will present his work on the choice of priors in
generalized cross entropy (GCE) models. To serve as a discussant,
please contact Daniel Kuehn at dan.p.kuehn@gmail.com.
Entropic Inference in
Physics and in Economics
The Info-Metrics Institute continues to provide us with a unique
interdisciplinary forum where we can extend our understanding
of entropic and Bayesian inference towards a unified framework
for processing information. In my field of physics, for example,
in the past these insights have provided a foundation for
thermodynamics and statistical mechanics. More recently, they
have led to a new foundation for other fundamental laws such as
Newton’s F = ma in classical mechanics, and the Schroedinger
equation in quantum mechanics. From this perspective the laws
of physics are no longer to be regarded as fundamental laws of
nature but rather as highly effective rules to process information
about the world – just like any other science, including economics.
In collaboration with Amos Golan we have recently developed an
entropic framework to model economies from the point of view of
an external observer who has very limited access to information
about the individual agents. The framework, which relies purely
on macroscopically accessible information, avoids the usual
rationality assumptions and leads to a different perspective on
the nature of economic equilibrium.
- Ariel Caticha (Professor and Chair, Department of Physics, University
at Albany - SUNY, Member of the Info-Metrics Advisory Board)
4
Info-Metrics Institute Fall 2014 Newsletter
Our Work – Leading Interdisciplinary Info-Metrics Research
Following is a list of representative research and recent publications by some of the Institute’s affiliates. More detailed information can be
found in the 2013-2014 Info-Metrics Annual Report, which will be available on our website in December 2014.
Pieter Adriaans (University of Amsterdam)
Current Research
• Developed a better understanding of facticity, especially its
relation to the well-known Kolmogorov structure function.
Both concepts appear to belong to a ‘landscape’ of meaning
functions.
• With the group of Luis Antunes in Porto, formulated new
results concerning safe approximations of Kolmogorov
complexity.
• COMMIT (http://www.commit-nl.nl) project: investigated
practical algorithms for the estimation of the information
content of very large knowledge graphs with the group of
Frank van Harmelen at the Free University in cooperation
with Elseviers Science Publishers. Extending this research
to the more general question of how the Kolmogorov
complexity of multidimensional objects and dimensionless
objects (like Graphs) can be estimated in a systematic way.
Hoping to test insights on the very large RDF graphs within
the next year.
Forthcoming/Recent Publications
• A safe approximation for Kolmogorov complexity, ALT
2014 (with P. Bloem, F. Mota, S. De Rooij, L. Antunes and
P. Adriaans)
• “Schilderen voor het brein” (Painting for the brain), in
Dutch, an English translation is under way, September
2013
Other Updates
• Invited Lecture and Seminar: Art, Information and Meaning,
McLuhan, University of Toronto, March 2014
• Public Lecture: Computers and Creativity, ASML Eindhoven,
May 2014
• We will have a concluding one week special conference
of the Atlas of Complexity project in Leiden, November
2014
Radu Balan (University of Maryland)
Forthcoming/Recent Publications
• “Excursions in Harmonic Analysis”, Volume 1 and 2, based
on the February Fourier Talk (FFT) 2006-2011 conferences,
http://dx.doi.org/10.1007/978-0-8176-8376-4 and http://
dx.doi.org/10.1007/978-0-8176-8379-5
• “Excursions in Harmonic Analysis”, Volume 3 and 4, based on
the FFT 2012-2014 conferences, will appear in 2015
• “Multi-window Gabor frames in amalgam spaces”, Math. Res.
Lett. 21 (2014), Number 1, pp. 1-15, (with J.G. Christensen,
I.A. Krishtal, K.O. Okoudjou, and J.L. Romero)
• “Invertibility and robustness of phaseless reconstruction”, in
press at Appl. Comp. Harm. Anal. (with Y. Wang)
Other Updates
• Invited Lecturer: Short course given at the Joint Mathematical
Meeting AMS-MAA 2015, San Antonio
• Working on a joint project with Amos Golan on the topic of
“Intrinsic Value of Information”
Marine Carrasco (University of Montreal)
Current Research
• Primary work in the areas of estimation and inference in Econometrics.
• Applies techniques from the inverse problems literature to
information-based estimation methods.
• With a PhD student Guy Tchuente, recently showed that regularized
versions of the Limited Maximum Likelihood estimator are
consistent in presence of heteroskedasticity and are efficient with
many weak instruments.
Recent Publications
• “Optimal Test for Markov Switching”, Econometrica, 2014, Volume
82, Number 2, 765-784. (with Liang Hu and Werner Ploberger)
• “On the asymptotic efficiency of GMM”, Econometric Theory,
2014, Volume 30, Issue 2, 372-406 (with J.P. Florens)
• “Asymptotic Normal Inference in Linear Inverse Problems”,
Handbook of Applied Nonparametric and Semiparametric
Econometrics and Statistics, Oxford University Press, February
2014 (with J.P. Florens and E. Renault)
Ariel Caticha (University at Albany – SUNY) Current Research
• Development of entropic and Bayesian methods as a unified
framework for processing information.
• Application of principles of entropic inference to the foundations
of physics (statistical mechanics, quantum mechanics and general
relativity). • Application of principles of entropic inference to economic
modeling, with Amos Golan.
Recent Publications
• “Towards an Informational Pragmatic Realism”, Mind and
Machines, Volume 24, pp. 37-70, 2014
• “Entropic Dynamics: and Inference Approach to Time and
Quantum Theory”, J. Phys: Conference Series, Volume 504 (2014)
012009
• “An Entropic framework for Modeling Economies”, Physica A.,
Volume 408, p. 149, 2014 (with A. Golan)
Min Chen (University of Oxford)
Current Research
• Fundamentals of visualization (theories, metrics and empirical studies).
• Visual analytics (model visualization and anomaly visualization).
• A variety of visualization techniques and applications.
Recent Publications
• “Visual multiplexing”, Computer Graphics Forum, 2014
• “Glyph-based video visualization for semen analysis”, IEEE
Transactions on Visualization and Computer Graphics, 2014
Other Updates
• Co-chair of EuroVis 2014, Swansea, UK
• Co-chair of IEEE VAST 2014, Paris, France
• Co-chair of Info-Metrics Institute Spring Workshop on Philosophy of
Information and Information Processing, Oxford, UK, March 2015
continues on page 5
www.american.edu/info-metrics
Jon Michael Dunn (Indiana University Bloomington)
Recent Publications
• “Guide to the Floridi Keys”, Essay Review of Luciano Floridi’s The
Philosophy of Information, MetaScience, 22, pp. 93-98.
• “The Third Life of Quantum Logic: Quantum Logic Inspired by
Quantum Computing”, Journal of Philosophical Logic, Volume
42, pp. 443-459, (with Lawrence Moss and Zhenghan Wang) (our
editor’s introduction to a special issue: Quantum Logic Inspired
by Quantum Computation)
• “On the Decidability of Implicational Ticket Entailment”, The
Journal of Symbolic Logic, Volume 78, pp. 214-236 (with K. Bimbó).
This last solved a problem open for 52 years, second on the Typed
Lambda Calculus and its Applications Open Problems List.
• Gave invited lectures at Logica 2013, Workshop on Non-classical
Epistemic Logics at the Czech Academy of Sciences Institute of
Philosophy, the Munich Center for Mathematical Philosophy,
and Carnegie-Mellon University.
• Visiting professor last spring in Philosophy at the University of
Pittsburgh
5
Duncan Foley (The New School for Social Research)
Current Research
• Generic model of social interaction, externalities and bifurcating
equilibria.
• The Edgeworth-Pareto ensemble and endogenous inequality of
wealth and income from market transactions.
• Models of behavior with bounded entropy (“rational inattention”).
• Statistical equilibrium and information in games, with David
Wolpert.
Forthcoming Publications
• Paper on “Varieties of Keynesianism” forthcoming in
International Journal of Political Economy
Other Updates
• Revision of intermediate microeconomics textbook with Sam
Bowles
• Working on new edition of Growth and Distribution with Tom
MIchl and Daniele Tavani
Other Updates
• As a Senior Visiting Fellow gave a seminar on “Logic, Information,
Computation”, Info-Metrics Institute, October 2013
• Attended Info-Metrics Conference: Information, Instability, and
Fragility in Networks, University of Colorado Boulder, November 2013
Bowen Garrett (Urban Institute, Health Policy Center)
Luciano Floridi (University of Oxford)
Recent Publications
• The best evidence suggests the effects of the ACA on employment
will be small http://www.urban.org/health_policy/url.
cfm?ID=413109 (with Robert Kaestner) • Redistribution under the ACA is modes in scope http://www.
urban.org/health_policy/url.cfm?ID=413023 (with S. Dorn and
J.Holahan)
• Midwifery care at a freestanding birth center: a safe and effective
alternative to conventional maternity care, Health Services
Research (with S. Benatar, E. Howell, and A. Palmer)
Forthcoming/Recent Publications
• “The Ethics of Biomedical Big Data” (2014-2015), funded by the
John Fell Main Award
• The Fourth Revolution – How the infosphere is reshaping human
reality, Oxford University Press, 2014
• Information Closure and the Sceptical Objection, Synthese,
Volume 191.6, pp. 1037-1050, 2014
• An Analysis of Information in Visualisation, Synthese, 190.16,
pp. 3421-3438, 2013 (with M. Chen)
• The Ethics of Information, Oxford University Press, 2013
• Distributed Morality in an Information Society, Science and
Engineering Ethics, Volume 19.3, pp. 727-743
• What is a Philosophical Question?, Metaphilosophy, Volume
44.3, pp. 195–221
Other Updates
• Member of the Académie Internationale de Philosophie des
Sciences (MAIPS), 2013
• Weizenbaum Award, by the International Society for Ethics and
Information Technology, for his “significant contribution to the
field of information and computer ethics, through his research,
service, and vision”, 2013
• Fellow of the British Computer Society (FBCS), 2013
• Cátedras de Excelencia Prize by the University Carlos III of Madrid, 2014
• Fernand Braudel Senior Fellowship by the European University
Institute, 2015
Featured Seminar of the Year:
J. Michael Dunn (Indiana University)
“Logic, Information and Computation”
http://www.american.edu/cas/economics/info-metrics/video.cfm
To see the full list of seminars, please visit our webpage.
Current Research
• Effects of the Affordable Care Act on labor market outcomes.
• Payment methods for health care services under Medicare and Medicaid.
• Policy and program evaluation.
Ramo Gencay (Simon Fraser University)
Forthcoming/Recent Publications
• A Chinese translation of my “Introduction to
High
Frequency
Finance”
book
with
China
Science Publishing & Media Ltd, scheduled to be published in
September 2015
• Jump detection with wavelets for high-frequency financial time
series, Quantitative Finance, forthcoming in 2014, (R. Gençay,
Y. Xue and S. Fagan)
• Private information and its origins in an electronic foreign
exchange market, Economic Modelling, Volume 33, pp. 86-93,
2013, (with R. Gençay and N. Gradojevic)
• Fuzzy logic, trading uncertainty & technical trading, Journal of
Banking and Finance, Volume 37, pp. 578-586, 2013 (with R.
Gençay and N. Gradojevic)
Other Updates
• Book
originally
published
in
2001
has
sold
over 8,000 copies worldwide and is available at:
http://www.amazon.com/exec/obidos/ASIN/0122796713/ref=
sr_1_2_1/103-1727768-4843068
• Invited Professor, University of Zurich, 2014
• Invited Professor, Bilgi University, 2013
continues on page 6
6
•
•
•
•
•
Info-Metrics Institute Fall 2014 Newsletter
Invited Professor, University of Chile, Santiago, Chile, 2013
Invited Professor, Karaganda State Technical University, 2013
Invited Professor, University of Navarra, 2013
Invited Professor, University of Zurich, 2013
Keynote Address: Economic links and counterparty risk,
International Istanbul Finance Congress, Kadir Has University,
Istanbul, Turkey, May 2013
John Geweke (University of Technology Sydney)
Current Research
• One of 22 chief investigators in the newly launched Centre of
Excellence for Mathematical and Statistical Frontiers of Big Data,
Big Models, New Insights. The Centre’s core funding is $20 million
from the Australian Research Council for the period July 2014
through June 2021, and is supplemented by more than $10 million
from the participating universities (University of Melbourne,
University of Technology Sydney, University of New South Wales,
and Queensland University of Technology.) Peter Hall of University
of Melbourne heads the Centre.
• Leading the development of the Sequential Analysis of Bayesian
Learning Algorithm (SABL) with support from a different ARC
grant, the enthusiastic participation of several colleagues, and
technical contributions by postdocs employed on the grant. A
Matlab toolbox of the same name will have its first release shortly.
• Continues as a half-time consultant for Amazon in Seattle, using
big data for forecasting and decision-making. 2013-14 marks
his final year as Theil Professor of Econometrics at Erasmus
University Rotterdam.
Forthcoming/Recent Publications
• “Analysis of Variance for Bayesian Inference”, Econometric
Reviews, Volume 33, pp. 270-288., 2014 (with G. Amisano)
• “Financial Competence, Risk Presentation and Retirement
Portfolio Preferences”, Journal of Pension Economics and
Finance, Volume 13, pp. 27-61, 2014
• “Improving Asset Price Prediction when All Models are False,”
forthcoming in Journal of Financial Econometrics (with G.
Durham)
• “Likelihood-based Inference for Regular Functions with
Fractional Polynomial Approximations,” forthcoming in Journal
of Econometrics (with L. Petrella)
• Invited review of the book: “Public Policy in an Uncertain
World,” forthcoming in Journal of Economic Literature
The Info-Metrics Institute thanks XacBank for
their support and interest in the Institute’s research.
XacBank is one of the leading financial institutions
in Mongolia. It maintains leadership positions in
microfinance, supporting women entrepreneurs,
providing financing to small and medium
enterprises, and promoting innovative financing
mechanisms for sustainable energy projects. We look
forward to years of working together.
Web link: http://www.xacbank.mn/
Other Updates
• Presentations in the last year include: Erasmus University,
University of Illinois, Midwest Econometric Meetings, European
Central Bank, Queensland University of Technology
• Effective August 1, 2014, my affiliation is the economics
department of Vanderbilt University
• Summary of info-metrics related work, John Geweke, July 2013
– June 2014
Amos Golan (American University)
Current Research
• The foundations of info-metrics.
• Value of information.
• Info-metrics modeling and inference.
• Markov Process and info-metrics.
• Priors and Information.
Recent Publications
• Information Dynamics in Minds and Machines, October 2013
• An entropic framework for modeling economies, Physica A. (with
A. Caticha)
Other Updates
• Invited Visiting Researcher: Santa Fe Institute, September 2013
and May 2014
• Visiting Professor: Faculty of Science and Department of
Economics, Hebrew University
• Six Day Advanced Info-Metrics Tutorial – Chiang Mai University,
Thailand
• Three Year Visiting Fellowship Appointment, St. Cross College,
Oxford
Alastair Hall (University of Manchester)
Current Research
• Info-metric approaches to inference about the parameters of
economic and statistical models based on the information in
moment conditions.
• Applications of info-metric methods to economic models
examining issues related to monetary policy, health expenditures
by the UK government, and the returns to education.
Recent Publications
• Testing for Structural Instability in Moment Restriction Models:
an Info-metric Approach, Econometric Reviews, forthcoming in
2014 (with A. Hall, Y. Li, C. Orme and A. Sinko)
• Inference in the Prescence of Redundant Moment Conditions
and the Impact of Government health Expenditure on Health
Outcomes in England, Econometric Reviews, forthcoming in
2014 (with M. Andrews, O. Elamin, A. Hall, K. Kyriakoulis and
M. Sutton)
Other Updates
• Attended the Econometrics Section of the Agricultural and
Applied Economics Association Conference (Joint with CAES),
Washington DC, August 2013
• Attended the Conference on Indirect Estimation Methods in
Finance and Economics, Abbey Hegne, Allensbach, Germany,
May 2014
• E.J. Hannan Lecture, Australasian Meeting of the Econometric
Society, Tasmania, Australia, July 2014
continues on page 7
www.american.edu/info-metrics
Heath Henderson (Inter-American Development Bank)
Forthcoming Publications
• “Structural transformation and smallholder agriculture:
An information-theoretic analysis of the Nicaraguan case”,
forthcoming in Agricultural Economics
Other Updates
• Accepted a teaching position at Iowa State University and will be
leaving the Inter-American Development Bank in the fall
• Attended the Complex Systems Summer School at the Santa Fe
Institute in New Mexico
Atsushi Inoue (Vanderbilt University)
Other Updates
• Now affiliated with the Economics Department of Vanderbilt
University
George Judge (University of California, Berkeley)
Current Research
• Stochastic dynamic economic systems.
• Non-linear Ordinal economic time series.
• Information recovery in binary networks.
• Causal path entropy.
Forthcoming/Recent Publications
• Information Theory model for time series prediction in time
series, Physica A, (with Plastino)
• Information recovery in binary network flow problems, Applied
Econometric Letters (with W. Cho)
• Generalizing Benford’s Law, Chapter in Theory and Application
of Benford’s Law, Princeton University Press (with J. Lee and W.
Cho)
Nicholas Kiefer (Cornell University)
Other Updates
• Keynote Lecturer: “Econometric Problems Arising in Banking”
at the China Meetings of the Econometric Society, June 2014
Taisuke Otsu (London School of Economics)
Forthcoming/Recent Publications
• Second-order refinement of empirical likelihood for testing
overidentifying restrictions, Econometric Theory, Volume 29,
pp. 324-353, 2013 (with Y. Matsushita)
• On testability of complementarity in models with multiple
equilibria, Economics Letters, Volume 120, pp. 79-82, 2013(with
Y. Rai)
• Robustness, infinitesimal neighborhoods, and moment
restrictions, Econometrica, Volume 81, pp. 1185-1201, 2013
(with Y. Kitamura and K. Evdokimov)
• Estimation and inference of discontinuity in density, Journal of
Business & Economic Statistics, Volume 31, pp. 507-524, 2013
(with K. Xu and Y. Matsushita)
• On Bartlett correctability of empirical likelihood in generalized
power divergence family, Statistics & Probability Letters, Volume
86, pp. 38-43, 2014 (with L. Camponovo),
• Empirical likelihood for regression discontinuity design,
forthcoming in Journal of Econometrics (with K. Xu and Y.
Matsushita)
• Robustness of bootstrap in instrumental variable regression,
forthcoming in Econometric Reviews (with L. Camponovo)
7
Other Updates
• Associate Editor of Journal of Business & Economic Statistics
• Associate Editor of Econometrics Journal
• European Research Council Consolidator Grant (SES-072961)
Jean-Pascal Nguessa Nganou (The World Bank - Uganda)
Current Research
• Exploring entropy-based methods for the estimation of fiscal
rules in resource-rich African countries.
• Entropy-based estimation of import elasticities for use in CGE
models.
• Using marginal analysis to determine the likelihood of a new
resource-rich country to experience a crowding out of nonresource government revenue caused by the revenue from
resource. This “moral hazard” issue is common in resource rich
countries.
• Econometric analysis of the role of institutions in the
determinants of Foreign Direct Investment (FDI) in low income
countries.
Rossella Bernardini Papalia (University of Bologna)
Current Research
• Entropy-based methods in ecological inference with spatial dependence.
• Small area estimation problems.
• Spatial econometric models for panel data.
Forthcoming/Recent Publications
• Information Theoretic competitiveness composite indicator at
micro level, revised and resubmitted for publication in Social
Indicators Research (with P. Calia and C. Filippucci)
• Developing a composite index by using spatial latent modeling
based on IT estimation, revised and resubmitted for publication
in Quality and Quantity (with E. Ciavolino)
• Trade costs in bilateral trade flows: Heterogeneity and zeroes
in structural gravity models, revised and resubmitted for
publication in The World Economy (with S. Bertarelli)
M. Hashem Pesaran (University of Cambridge)
Forthcoming/Recent Publications
• “Optimal Forecasts in the Presence of Structural Breaks”,
Journal of Econometrics, Volume 177, Issue 2, pp.134-152,
December 2013, http://www.sciencedirect.com/science/article/
pii/S0304407613000687 (with A. Pick and M. Pranovich)
• “An Empirical Growth Model for Major Oil Exporters”, Journal
of Applied Econometrics, Volume 29, Issue 1, pp. 1-21, January/
February 2014, http://onlinelibrary.wiley.com/doi/10.1002/
jae.2294/abstract (with H.S. Esfahani and K. Mohaddes)
• “Aggregation in Large Dynamic Panels”, Journal of Econometrics,
Volume 178, Issue 2, pp. 273-285, January 2014. http://www.
sciencedirect.com/science/article/pii/S0304407613001917 (with
A. Chudik)
• “Signs of Impact Effects in Time Series Regression Models”,
Economics Letters, Volume 122, Issue 2, pp.150-153, February
2014,http://www.sciencedirect.com/science/article/pii/
S0165176513005028 9 (with R. Smith)
• “Testing Weak Cross-Sectional Dependence in Large Panels”,
January 2012. CWPE Working Paper No. 1208, IZA Discussion
Paper No. 6432, forthcoming in Econometric Reviews
continues on page 8
8
•
•
•
Info-Metrics Institute Fall 2014 Newsletter
“Constructing Multi-Country Rational Expectations Models”,
CESifo Working Paper No. 3081, October 2013, forthcoming
in Oxford Bulletin of Economics and Statistics (with S. Dees, R.
Smith and L.V. Smith)
“Large Panel Data Models with Cross-Sectional Dependence:
A Survey”, CESifo Working Paper No. 4371, August 2013,
forthcoming in B. H. Baltagi (Ed.), The Oxford Handbook on
Panel Data. Oxford University Press (with A. Chudik)
“Common Correlated Effects Estimation of Heterogeneous
Dynamic Panel Data Models with Weakly Exogenous Regressors”,
July 2014, IZA Discussion Paper No. 6618, forthcoming in the
Journal of Econometrics (with A. Chudik)
Werner Ploberger (Washington University St. Louis)
Recent Publications
• “Optimal Test for Markov Switching Parameters”
• By analyzing the likelihood function, we show that the testing
problem for switching parameters is asymptotically equivalent to
a Gaussian shift with a non-standard rate, Econometrica, April
2014 (with M. Carrasco and L. Hu)
Other Updates
• Awarded Fernand-Braudel fellowship for a 3-month visit at the
European University Institute
Jeffrey Racine (McMaster University)
Current Research
• Smooth nonparametric quantile estimation.
• Nonparametric instrumental variable estimation.
• Shape constrained nonparametric instrumental regression.
• Computationally efficient nonparametric estimation.
Forthcoming/Recent Publications
• “Infinite-Order Cross-Validated Local Polynomial Regression”,
forthcoming in the Journal of Econometrics (P. Hall and J.S.
Racine)
• “Spline Regression in the Presence of Categorical Predictors”,
forthcoming in the Journal of Applied Econometrics (S. Ma, J.S.
Racine and L. Yang)
• “A Partially Linear Kernel Estimator for Categorical Data”,
forthcoming in the Econometric Reviews (Q. Gao, L. Liu and J.S.
Racine)
• “Mixed Data Kernel Copulas”, forthcoming in Empirical
Economics (J.S. Racine, J.S.)
• Racine, J.S. and C. Parmeter (2014), “Data-Driven Model
Evaluation: A Test for Revealed Performance”, in `Handbook of
Applied Nonparametric and Semiparametric Econometrics and
Statistics’, Oxford University Press, (A. Ullah, J.S. Racine and L.
Su Eds) pp. 308-345
Sherman Robinson (Environment and Production Technology
Division, Development Strategies and Governance Division,
International Food Policy Research Institute)
Recent Publications
• Estimating Parameters and Structural Change in CGE Models
Using a Bayesian Cross-Entropy Estimation Approach (with D.
Go, H. Lofgren, F. Ramos)
Other Updates
• Presented a paper for the GTAP conference (Global Trade and
Analysis Project) in Dakar, Senegal, June 2014
Teddy Seidenfeld (Carnegie Mellon University)
Recent Publications
• “The Effect of Exchange Rates on Statistical Decisions”,
Philosophy of Science, Volume 80, Number 4, pp. 504-532, 2013
(with M.J.Schervish and J.B.Kadane)
• “Two Theories of Conditional Probability and NonConglomerability”, ISIPTA-13 Proceedings of the Eighth
International Symposium on Imprecise Probabilities and
Their Applications, 2013 (with F. Cozman, T. Denoeux, T., S.
Destercke, and T. Seidenfeld); (eds.) SIPTA (with M.J.Schervish
and J.B.Kadane)
• “Dominating Countably Many Forecasts,” Annals of Statistics,
Volume 42, Number 2, pp. 728-756, 2014 (with M. Schervish
and J. Kadane)
Richard J. Smith (Cambridge University)
Current Research
• A major research theme concerns the general theory of
estimation and inferential methods for econometric models
specified through moment conditions. The emphasis is on
alternatives to the standard generalized method of moments,
in particular, those within the generalized empirical likelihood
class. The design of valid inferential methods when confronted
by instruments weakly correlated with endogenous variables is a
recent theme and currently this research has dealt with models
specified by conditional moment restrictions.
• Current projects include the development of reliable inferential
methods in the moment condition context based on the bootstrap
and neglected heterogeneity in moment condition models.
Forthcoming/Recent Publications
• Refereed Published Paper: “Neglected Heterogeneity in Moment
Condition Models”, Journal of Econometrics (2014), 178, 86-100
(with J. Hahn and W.K. Newey)
• Refereed Paper Accepted for Publication: “Recent Developments
in Empirical Likelihood and Related Methods”, forthcoming in
Annual Review of Economics (with P.M.D.C. Parente)
• Paper Under Revision for Resubmission: “Additional Conditional
Moment Tests” previously “Exogeneity in Semiparametric
Models: Definitions and Tests” (with P.M.D.C. Parente)
Other Updates
• International Keynote Speaker: Statistics Day Conference,
Reserve Bank of India, Mumbai, India
• Distinguished Speaker: Singapore Economics Review
Conference, Singapore
• National Seminar: University of Surrey; University of Cambridge.
• International Seminar: CEMFI/UC3 Madrid Econometrics
Seminar; University of Melbourne; Reserve Bank of India,
Mumbai, India; Nanyang Technical University, Singapore
• Professional Activities: Chair of the Faculty of Economics,
University of Cambridge
• British Academy: Section S2 Committee, Section S2 Small
Research Grants Panel Chair
• Royal Economic Society: Managing Editor, The Econometrics
Journal; Executive Committee, Royal Economic Society: Council,
Royal Economic Society; Journals Publication Subcommittee,
Royal Economic Society
• UK Government Statistical Service: Methodology Advisory
Committee
continues on page 9
www.american.edu/info-metrics
•
•
•
•
•
•
•
•
•
Econometric Foundation: Originating Council
Econometric Theory: Advisory Board
EC2: Standing Committee
Info-Metrics Institute: Advisory Board
Advisory Commission: CEFAGE-UE, Universidade de Evora
Portugal
Governor: National Institute of Economic and Social Research.
Centre Fellow: Centre for Microdata Methods and Practice,
U.C.L. and I.F.S.
Visiting Fellow: National Institute for Economic and Social
Research
Co-Editor: Themes in Modern Econometrics (Cambridge
University Press)
Ehsan Soofi (University of Wisconsin Milwaukee)
Recent Publications
• “Comparison, Utility, and Partition of Dependence under
Absolutely Continuous and Singular Distributions”, Journal
of Multivariate Analysis, 2014 (with N. Ebrahimi and N.Y.
Jalali)
• “New Maximum Entropy Methods for Modeling Lifetime
Distribution”, Naval Research Logistics, 2014 (with M. Asadi,
N. Ebrahimi, and S. Zarezadeh) Michael Stutzer (University of Colorado Boulder)
Current Research
• Newly proposed entropy bounds on asset pricing model kernels.
Forthcoming/Recent Publications
• “Optimal Hedging Via Large Deviations”, Physica A.
Aman Ullah (University of California Riverside)
Forthcoming/Recent Publications
• Nonparametric Regression Estimation with General Parametric
Error Covariance: A More Efficient Two-Step Estimator,
Empirical Economics, Volume 45, pp. 1009-1024, 2013 (with Su
and Wang)
• Parametric and Nonparametric Frequentist Model Selection and
Model Averaging, Econometrics, Volume 1, Number 2, pp. 157179, 2013 (with H. Wang)
• On Existence of Moment of Mean Reversion Estimator in Linear
Diffusion Models, Economics Letters. (2013), 120, 146-148 (with
Bao and Zinde-Walsh)
• Handbook of Applied Nonparametric and Semiparametric
Econometrics and Statistics, New York: Oxford University Press,
2014 (with J. Racine and L. Su)
• Robustified Financial Time Series Forecasting with Bagging,
Econometric Reviews, Volume 33, pp. 575-605, 2014 (with S. Jin
and L. Su)
• Nonparametric and semiparametric regressions subject to
monotonicity constraints: Estimation and forecasting, Journal of
Econometrics, 182, 196-210, 2014 (with Lee and Tu)
• Testing Additive Separability of Error Term in Nonparametric
Structural Models, forthcoming in Econometric Reviews (with
Su and Tu)
• A Semiparametric Conditional Duration Model, forthcoming in
Economics Letters, (with Dungy, Long, Wang)
• Moment Approximation for Unit Root Models with Nonnormal
Errors, forthcoming in Advances in Econometrics (with Bao and
Zhang)
9
David Wolpert (Santa Fe Institute)
Current Research
• The information geometric structure underlying all
noncooperative games.
• Modeling cyber-security scenarios with game theory.
• Modeling UAV / air tanker fire-fighting scenarios with game
theory.
• Applying information theory to Theory of the Firm.
• Solving for the optimal macrophysics to describe a given
microphysics.
• Beating the Landauer bound.
• A semantic theory of information and the foundations of
physics.
Forthcoming/Recent Publications
• “Cyber-Physical Security: A Game Theory Model of Humans
Interacting over Control Systems”
•
“Game Mining: How to Make Money from those about to Play
a Game” • “Application of game theoretic models to evaluate airline equipage dynamics of Nextgen technologies”
• “A theory of unstructured bargaining using distribution-valued
solution concepts” • “Towards a Bayesian network game framework for evaluating
DDoS attacks and defense”
• “Predicting the behavior of interacting humans by fusing data
from multiple sources”
• “Counter-Factual Reinforcement Learning: How to Model Decision-Makers That Anticipate the Future”
• “Hysteresis effects of changing parameters of noncooperative
games”
Other Updates
• Moved full-time to Santa Fe Institute
Ximing Wu (Texas A&M University)
Current Research
• Nonparametric and information-theoretic methods in
econometrics.
• Data-driven information-theoretic methods of distributional
and specification hypotheses.
• Nonparametric estimation of multivariate density and copula
density functions.
• Semiparametric
estimation
of
shape-constrained
functions.
Forthcoming/Recent Publications
• “Smoothing Tests of Copula Specifications”, forthcoming in
Journal of Business and Economic Statistics (with J. Lin)
• “Data-driven Smooth Test of Symmetry”, Journal of Econometrics
(with Q. Li and D. Zhang)
• “Penalized exponential series estimation of copula densities with
an application to intergenerational dependence of body mass
index”, forthcoming in Empirical Economics (with Y. Gao and
Y. Zhang)
• “Interdependence of oil prices and stock market indices: A
copula approach”, Energy Economics, Volume 44, pp. 331–339,
2014 (with K. Sukcharoen, T. Zohrabyan, D. Leatham)
continues on page 10
10
Info-Metrics Institute Fall 2014 Newsletter
Victor Yakovenko (University of Maryland)
Current Research
• Studying probability distributions and inequality in money,
income, debt, and energy consumption per capita around the
world.
Forthcoming/Recent Publications
• “Global inequality in energy consumption from 1980 to 2010”,
Entropy, Volume 15, pp. 5565, 2013, open access http://dx.doi.
org/10.3390/e15125565; http://arxiv.org/abs/1312.6443 (with S.
Lawrence, Q. Liu, and V. M. Yakovenko)
• Continuing research grant from the Institute for New Economic
Thinking (INET) for the period of 1/2013 - 12/2014 on the topic
“Statistical Physics Approach to Income and Wealth Distribution”
http://ineteconomics.org/grants/statistical-physics-approachincome-and-wealth-distribution
Other Updates
• Econophysics materials are regularly posted on a webpage:
http://physics.umd.edu/~yakovenk/econophysics/
• Coverage by the media: Press release “The Entropy of Nations”
http://jqi.umd.edu/news/entropy-nations
• Coverage by the media: TV interview on Brian Lehrer’s show in
New York http://www.cuny.tv/show/brianlehrer/PR2002638
• Coverage by the media: Appearance on TV Science
Channel in the episode “Is Poverty Genetic?” of the
program “Through the Wormhole with Morgan Freeman”
http://www.sciencechannel.com/tv-shows/through-thewormhole/videos/the-rich-dont-play-by-the-same-rules.htm
• Coverage by the media: Science magazine coverage “Physicists
say it’s simple” the special section “The Science of Inequality”,
http://www.sciencemag.org/content/344/6186/828
• Invited econophysics talks given at: University of Illinois at
Urbana-Champaign, Physics Department Colloquium, August
2013
• Invited econophysics talks given at: University of Maryland,
CSCAMM/KI-Net Seminar, September 2013
• Invited econophysics talks given at:American University,
Washington DC, Physics Colloquium, February 2014
• Invited econophysics talks given at: Texas A&M University,
College Station, Physics Department Colloquium, March 2014
• Invited econophysics talks given at: Rutgers University,
Piscataway, Physics Department Colloquium, April 2014
• Invited econophysics talks given at: Graduate School of the City
University of New York (CUNY), May 2014
• Invited econophysics talks given at: New School for Social
Research, New York, Department of Economics Seminar, May
2014
• Invited econophysics talks given at: Institute for New Economic
Thinking (INET), New York, May 2014
Fall 2013/Spring 2014
Visiting Fellows
J. Michael Dunn (U Indiana Bloomington)
Ariel Caticha (SUNY Albany)
Esfandiar Maasoumi (Emory U)
Aman Ullah (UC Riverside)
Yiping Yang (Beijing University)
Forthcoming Summer 2015 Tutorials
“Nonparametric Kernel Methods for Practitioners
across the sciences”
Jeffrey S. Racine (McMaster University)
“Interdisciplinary Applications of Microeconometrics.”
William H. Greene (New York University)
For further information on these tutorials, please visit our
Info-Metrics Summer Program page: http://www.american.
edu/cas/economics/info-metrics/econometrics.cfm
The Office of the Comptroller of the Currency
(OCC) is the primary sponsor of the Info-Metrics
Institute. We wish to thank OCC for their
continued generous support. We would like to
thank XacBank Mongolia for help in supporting
the Information, Instability and Fragility in
Networks: Methods and Applications conference
in November 2013. We thank the American
University College of Arts and Sciences as well
as the Economics department for partial support
toward general Institute activities.
Participants get ready for the Information, Instability and Fragility
workshop in November of 2013 at Hotel Boulderado in Boulder, CO
In order to encourage collaborations among its
Affiliates, the Info-Metrics Institute makes
available limited funds to partially support those
Affiliates (Research Associates and Advisory Board
Members) that are either currently collaborating or
interested in collaborating (meaning at least two Affiliates work together on a project/paper) and where such
funding would facilitate the collaboration. For more
information, please email info-metrics@american.edu.
www.american.edu/info-metrics
11
Affiliate Profiles
Marine Carrasco
Marine Carrasco (Professor at the
University of Montreal) works
primarily in the areas of estimation
and inference in Econometrics. She
applies techniques from the inverse
problems literature to informationbased estimation methods. With a
PhD student Guy Tchuente, she recently showed that regularized
versions of the Limited Maximum Likelihood estimator are
consistent in presence of heteroskedasticity and are efficient
with many weak instruments. Her latest research is focused on
the generalization of empirical likelihood estimation to handle a
continuum or a countable infinite number of moment conditions
and the regularization techniques of the covariance matrix and its
application to portfolio selection in finance.
Professor Carrasco’s progressive research on the asymptotic efficiency of
GMM was done jointly with Jean-Pierre Florens and will be forthcoming
in Econometric Theory. Her work with Guy Tchuente on the efficient
estimation with many weak instruments using regularization techniques
and regularized LIML for many instruments can be found at: https://
www.webdepot.umontreal.ca/Usagers/carrascm/MonDepotPublic/
carrascm/EconRev.pdf and https://www.webdepot.umontreal.ca/
Usagers/carrascm/MonDepotPublic/carrascm/Carrasco_Tchuente.pdf.
Min Chen
Min Chen (Professor of Scientific
Visualization at Oxford University
and a Fellow of Pembroke College)
developed his academic career in
Wales between 1984 and 2011. He
is leading visualization activities at
Oxford e-Research Centre, with a
team of researchers working on a
broad spectrum of interdisciplinary research topics, ranging from
sciences to sports, and from digital humanities to cybersecurity. He
is particularly passionate about developing a theory of visualization
that can explain why and how good visualization facilitates efficient
and effective information delivery from computers to humans
and between humans. In collaboration with several colleagues, he
proposed to use information theory as an underpinning framework
for visualization (2010); he related the merits and demerits of visual
embellishment to source and channel encoding in information
theory (2012); and more recently, he proposed 10 types of visual
multiplexing, categorizing how humans perceive multiple pieces of
information associated to a single spatial location (2014). In aspects
of theory development, he is currently working on informationtheoretic metrics for measuring quality of visualization (QoV).
Min Chen is a fellow of British Computer Society, European
Computer Graphics Association, and Learned Society of Wales.
For more information about his research interests, publication and
teaching activities, please visit his web page (https://sites.google.com/
site/drminchen).
Nathan Harshman
Nathan Harshman (Associate Professor
of Physics, American University) applies
techniques from quantum information theory
to particle systems as a method to probe
the limits of deterministic and reductionist
descriptions of matter. His research explores
whether information-based criteria can be
employed to discover physically-important quantum subsystems and to
investigate the emergence of coherent quantum composite structures.
He work has shown how the geometrical and dynamical symmetries
of a system can privilege certain observables, and these observables
can in turn induce virtual subsystems. The entanglement among these
distinguished subsystems obeys a kind of informational conservation law.
Recently, he presented applications of these ideas to few-body physics. At
the Workshop on Noise, Information and Complexity at the Quantum
Scale in Erice, Italy, at the Workshop on Universality in Few-Body Systems
at the Institute for Nuclear Theory at the University of Washington, and
at the Annual Meeting of the Division of Atomic, Molecular and Optical
Physics of the American Physical Society. In these presentations and two
recent articles in Physical Review A, he described entanglement structures
that present new opportunities for implementing quantum information
processing and quantum simulations with highly-controllable particle
systems like ultracold atomic gases. Previous related research focused on
classifying entanglement structures for two-body systems by how they
transform under dynamical symmetries and on the general properties of
entanglement relativity under redefinition of quantum subsystems.
Jeffrey Racine
Jeffrey S. Racine (Senator William McMaster
Chair in Econometrics at McMaster University)
works in econometrics and statistics. His main
field of interest lies in robust estimation and
inference and he works mainly with kernelbased and spline-based approaches. He is
most well-known for his work on robust
estimation in the presence of categorical and continuous variables. His
current research includes exploring the links between kernel smoothing
methods and hierarchical Bayes methods (joint with Professor Nick Kiefer
at Cornell University), fully nonparametric approaches to instrumental
variable methods (one project joint with Professor Xiaohong Chen at Yale
University, another joint with Professor Jean-Pierre Florens at the University
of Toulouse), and nonparametric approaches with improved performance
(joint with Professor Peter Hall at the University of Melbourne).
Professor Racine’s recent travels have taken him to CREATES at the University
of Aarhus, Denmark (Center for Research in Econometric Analysis of Time
Series), where he is an International Research Fellow, The University of
Sydney, the University of Melbourne, Monash University and La Trobe
University in Australia where he gave seminars and lectures, and recent
conferences at the Rimini Center for Economic Analysis held in Rimini Italy
(Bayesian Workshop) where he is also a Senior Fellow, the second ISNPS
conference held in Cadiz, Spain (International Society for Nonparametric
Statistics where he is on the organizing and charting committees), and
IWAP 2014 (International Workshop on Applied Probability).
12
Info-Metrics Institute Fall 2014 Newsletter
Graduate Students in Their Own Words
Paul Corral (American University)
Most endogenous sample selection issues have been dealt with
variants of Heckman’s selection model. The model relies on the
assumption of normality. However, when dealing with data where
the data generating process is unknown, and the data may be less
than ideal an information theoretic approach may be more suitable.
During the summer I have been working on implementing the
methodology previously developed in Golan, Moretti, and Perloff
(1999) to study the effects of liquidity constraints on agriculture
among Nigerian farming households.
Justin Grana (American University)
I am developing a method using generalized maximum entropy to
estimate the level of player sophistication in level-k games. In the
level-k framework, players that are level 0 do not act strategically
and choose their actions randomly. Level 1 players choose their
best action assuming all other players are level 0. Level 2 players
choose their best action assuming all other players are level 1...and
so on. Estimating parameters of this game (including the level of
reasoning of the players) is generally an under-determined problem
when no restrictions are placed on the level 0 distribution. Therefore,
most models assume the level 0 distribution is uniform over the
action space. My method relaxes this assumption and allows the
econometrician to estimate the level of player sophistication, as
well as other parameters of the game without assuming a uniform
level 0 distribution. Initial Monte Carlo experiments shows that
my method performs at least as well as the traditional maximum
likelihood methods. Ellis Scharfenaker (New School for Social Research)
Adviser: Duncan Foley
It is widely recognized in economic theory that competition plays
a central organizing role in the economy. However, it is quite
reasonable to assume that firms have a limited capacity to optimally
respond to the myriad of market signals. Therefore, in considering
the behavior and actions of firms in a decentralized competitive
landscape, it is important to explicitly consider the information
processing constraints as in Sims (2003). From this perspective I am
developing a field model of competitive firms that are constrained by a
minimum informational entropy. It shows that from the information
theoretic perspective, firms’ expected payoff from competing in
markets subject to a constrained Shannon information channel,
gives rise to a decision temperature that characterizes a scalar field
of competitive pressure over which firms’ potential entry and exit
decisions are distributed. This model is general and parsimonious
and is well supported by firm-level data gathered from Compustat.
Support the Institute!
The Info-Metrics Institute is happy to receive donations toward its different
activities. Contributions to the Institute are tax deductible, subject to federal
and state guidelines. With these resources, we hope to establish more longterm fellowships for students and junior and senior researchers. We also hope
to be able to expand our classes and knowledge dissemination activities.
For more information on how to donate, please contact the Institute at
info-metrics@american.edu or 202-885-3770.
Daniela Scida (Brown University)
Adviser: Eric Renault
The study of financial phenomena like contagion involves the use of
models with a large number of variables. These models are challenging in
that the presence of so many variables introduces “noise” in the assessment
of how “financial individuals” are connected and influence each other. My
line of research for the Info-Metrics summer fellowship is in the spirit of
sparse Vector Autoregressive modeling by Davis, Zang, and Zheng (2012),
where the end goal is to deliver more parsimonious models by deleting
weak links. However, in contrast with their paper, I attempt to achieve
some degree of sparsity by taking advantage of the asymmetric feature of
Granger Causality. Mehdi Shoja (University of Wisconsin Madison)
Adviser: Ehsan Soofi
Each quarter Central Banks survey panels of experts and collect probability
distributions for the pre-assigned categories of an economic variable, such
as GDP and consumer price index. Recently, Shannon entropy has been
used for measuring the uncertainty of the economic variable based on the
forecasters’ distributions. In this paper, my research team develops this line
of research by utilizing the information theoretic and Bayesian machineries.
Forecasters’ distributions are viewed as post-data distribution generated in
a random environment, which provide information about the economic
variable in terms of expected uncertainty reduction. The maximum entropy
probability vectors derived based on the moments of the forecasters’
distributions provide upper bounds for the uncertainty. Disagreement
among the forecasters is defined in terms of information divergence.
Bayesian models are used to capture the uncertainty about the unknown
probability distribution of the economic variable.
Events Over the Past 5 Years
Below is a list of the workshops and conferences the Institute has
thoughtfully organized over the past five years. For more information on
past conferences and workshops, visit the event archives on our website.
Information, Instability and Fragility in Networks: Methods and
Applications, University of Colorado, Boulder
Philosophy of Information, American University
Info-Metrics and Nonparametric Inference, University of California, Riverside
Information and Econometrics of Networks, American University
Information Theory and Shrinkage Estimation, American University
Philosophy of Information, American University
Info-Metrics Across the Sciences, American University
Info-Metrics: Theory and Applications in the Social Sciences,
American University
Info-Metrics in the Natural Sciences and Its Implications for the
Social Sciences, American University
Information Theoretic Estimation and Data Analysis: State of the
Science and New Directions, American University
Editors: Aisha Khan and Carol Hong.