Info-Metrics Institute
College of Arts and Sciences, American University
Annual Report
2011 - 2012
Table of Contents
Welcome
Message from the Director………………………………………………………………………...4
Message from the Office of the Comptroller of the Currency…………………………………….5
In Memoriam: Halbert White (1950-2012)…………………………………………..……………6
Notes from the Advisory Board……………………………...……………………………………7
Graduate Students: In Their Own Words…...……………………………………………………..9
Thanks to our Sponsors……………..……………………………………………………………10
Who We Are
Photo Album…………………………………………………………………………………..…11
2011 – 2012 Advisory Board…………………………………………………………………….14
2011 – 2012 Visiting Fellows …………………………………...………………………………16
Affiliate Profiles…………….……………………………………………………………………17
What We Do
Current Research



















Pieter Adriaans (University of Amsterdam)……………………………………………..19
Rossella Bernardini Papalia (University of Bologna)……………………..…………......19
Mehmet Caner (North Carolina State University)……………………………………….19
Marine Carrasco (University of Montreal………………………………………………..20
Luciano Floridi (University of Hertfordshire)……...……………………………………20
Duncan Foley (New School for Social Research and Santa Fe Institute)………………..20
A. Bowen Garrett (McKinsey & Co., and Urban Institute)………………..…………….20
Ramo Gencay (Simon Fraser University)………………………………………………..21
Amos Golan (American University)……………………………………………………..21
Alastair Hall (University of Manchester)………………………………………………..21
Atsushi Inoue ( NC State University)……………………………………………………21
George Judge (University of California-Berkeley)………………………………………22
Nicholas Kiefer (Cornell University)…………………………………………………….22
Robin Lumsdaine (American University)………………………………………………..22
Essie Maasoumi (Emory University)…………………………………………………….22
Ilya Nemenman (Emory University)………...…………………………………………..22
Jeffrey R. Racine (McMaster University)………………………………………………..23
Eric Renault (Brown University)..……………………………………………………….23
Richard Smith (University of Cambridge)……………………………………………….24
2
 David H. Wolpert (NASA Ames Research Center)……………………………………...24
 Ximing Wu (Texas A & M University)………………………………………………….24
 Victor Yakovenko (University of Maryland)……………………………………………25
2012 Summer Classes……………………………………………………………………………26
Events and Activities
 Institute Seminars………………………………………………………………………...33
 Workshops and Conferences……………………………………………………………..34
 Forthcoming Workshops and Conferences…………………………….………………...45
For More Information
 Institute Support………………………………………………………………………….49
 Website and Contact……………………………………………………………………..50
3
November 2012
Message from the Director
In the three short years since the inauguration of the Info-Metrics Institute, we have expanded
our activities and research beyond our expectations.
We have organized and sponsored seven conferences and workshops, each one exploring a
specific topic within info-metrics, including basic information-theoretic estimation; information
processing and inference in the natural sciences; information and information processing in
networks; basic philosophical meaning of information and information processes; and big data.
We ran seven summer tutorials (hands-on classes), which we expanded in the last year with a
more cross-disciplinary emphasis. These initiatives have helped establish a common language to
link disciplines in solving similar info-metric problems.
In the next year, we plan to expand our activities in three different ways: First, we will make a
special effort to reach young researchers and researchers from the natural, medical and
engineering sciences. Second, we will initiate activities outside the Washington, D.C. area, such
as the forthcoming November conference in Riverside. Third, we will work on improving the
interdisciplinary nature of the tutorials by utilizing multiple instructors for each workshop (as we
did with the information and data mining class provided last summer).
With these milestones achieved, and with the establishment of the necessary infrastructure, we
are also ready to move to the next phase of pushing the research frontiers of info-metrics and of
building our role as a leading interdisciplinary institute. In the next year, we will engage in an
open discussion among our affiliates, and others, about our overall goal and long term direction.
Please bookmark our website and check in frequently to find information about our activities.
We are always open to new ideas and for new initiatives.
As always, we thank the Office of the Comptroller of the Currency for its continued financial
support. With this assistance, we are carrying the most important objectives of the Institute:
helping to shape info-metrics research and future generations of researchers.
- Amos Golan, Director, Info-Metrics Institute
4
Message from the Office of the Comptroller of the Currency:
How Info-Metrics Helps the OCC
The Office of the Comptroller of the Currency (OCC) has supported the Info-Metrics Institute
since its creation. The Institute promotes research and scholarly interaction in areas of modeling
and analysis that directly apply to the bank supervision work of the OCC. With insights gained
in part through our association with the Institute, OCC staff is better able to evaluate the methods
used by banks, to suggest new approaches, and to encourage better practices.
In addition to oversight of bank modeling and analysis, the OCC has ongoing analytical needs of
its own. The OCC has significantly expanded its collection of highly granular, standardized data
from banks. Effective use of such data could fundamentally change the practice of bank
supervision, but that in turn requires development of appropriate methods of analysis and
presentation. OCC staff members have incorporated approaches based in information theory into
their own research, in part by drawing on the insights of scholars associated with the Institute.
Finally, the Institute is in a position to foster broader methodological improvements that could
make a material difference in the operation of banks and other financial firms. Improvements in
financial modeling play an important role in making the global financial system safer and less
susceptible to failure. The OCC has been pleased to support the development and growth of the
Institute, in view of its potential for broad and substantial benefits to banks, the financial system,
and the public.
-Mark Levonian, Senior Deputy Comptroller, Office of the Comptroller of the Currency
5
In Memoriam: Halbert White (1950-2012)
Halbert (Hal) White, an active member of the Info-Metrics Institute
and a member of its Advisory Board, passed away on March 31,
2012, following a four-year battle with cancer.
Hal was a true scholar, one of the world's leading econometricians,
and a kind and humble person. His numerous contributions in
econometrics include original and path-breaking work in, among
other topics, neural networks, entropy and information, asymptotic
theory, semi-parametric methods, causality, and robustness and misspecification issues.
His 1980 Econometrica paper, suggesting a method for computing
robust standard errors and a related test statistic, made him a
household name among the economics profession. Now commonly
referred to as “White standard errors”, this brilliant contribution is just one example of how he
changed the direction of empirical work. He will be greatly missed.
Esfandiar (Essie) Maasoumi, Hal’s good friend and colleague, penned a touching tribute in his
memory. Read the article on our website at http://www.american.edu/cas/economics/infometrics/white-obit-maasoumi.cfm.
6
Notes from the Advisory Board
The past year has been fruitful for the Info-Metrics Institute. We
organized two workshops on the philosophy of information (one at the
Institute and one as a special session at the centennial Turing
conference in Cambridge). There were also interesting developments
on the theoretical front.
The Fall 2011 workshop, Philosophy of Information, brought together
many researchers representing a broad range of disciplines, varying,
amongst others, from philosophy to physics and statistics, to computer
science, complexity and the social sciences.
The philosophy of information, a natural component of info-metrics, is also rapidly gaining
momentum, but as a young discipline it goes through a process of defining its central subject
matter and core research questions. Three issues, in my view, are important in this respect:
1) The notion of information should be studied in the context of its various rigorous
mathematical formulations. Any attempt to treat information as a general philosophical
notion, without any formal exactness, immediately collapses into the traditional
conundrums of epistemology and metaphysics;
2) The notion of information has deep historical roots that go back to the pre-Socratics; and
3) Theoretical complexity theory, as it emerged in the second half of the twentieth century,
sheds new light on important philosophical problems, notably induction.
The Info-Metrics Institute is an ideal environment to discuss and develop these ideas and the
philosophical connections between information, information processing and complexity. We
look forward to continuing those discussions and furthering related research in the coming years.
-Pieter Adriaans, Member
What a year! In the 2011-2012 Academic Year, the Info-metrics
Institute generated a remarkable amount and range of activity. By any
measure, we had an extraordinarily successful year. From staging
conferences, to hosting scholars for seminars and research
collaborations, and to delivering advanced material to graduate students
and professionals, the Institute confirmed its niche as the go-to place
for advancing and disseminating insights on info-metrics. One of the
most satisfying aspects of the Institute‟s mission has been to
demonstrate that info-metrics is a multidisciplinary enterprise. The
Institute provides a rare venue for genuine and fruitful interactions across disciplines.
The extent of participation in our activities reached new highs. Scholars came from far and wide,
from outstanding universities and government agencies to learn more about developments in
info-metrics and how they apply across several disciplines. We attracted 67 participants to our
7
October 2011 Philosophy of Information workshop, followed by 83 to our November 2011
Shrinkage Estimation workshop, and then topped by 146 participants at our March 2012
Information and Econometrics of Networks workshop. In addition, we made significant progress
in developing our website and in involving PhD students. For a small Institute, this record is
extraordinary. For the coming year, we plan to accomplish more in disseminating interesting
findings from our workshops, conferences, and research activities.
We are proud of our efforts so far to establish a home for scholars and others eager to innovate
and learn about info-metrics. At the same time, we look forward to a future that extends the field
and communicates the valuable and continuing contributions of info-metrics.
- Robert Lerman, Chair
Even though the interdisciplinary approach is one of the appealing
features of the Institute, I am glad, as a financial econometrician, to
report how much the Institute research agenda is topical in my field.
While Mark Levonian, Senior Deputy of the Office of the Comptroller of
the Currency (OCC) was emphasizing the importance of control of
“model risk” (see fall 2011 Newsletter of Info-Metrics institute), this
concern underlies many recent developments of modern econometrics
that are at the core of our research agenda. Under the general terminology
of “ill-posed inverse problems”, econometricians include various needs
of inference regarding infinite dimensional objects with limited
information. This implies dangerous discontinuities in calibration
procedures that make the model risk even more detrimental. An
information-based approach is typically well-suited as compellingly exemplified in recent
conferences of the Institute about shrinkage and nonparametric approaches. Applications to
financial econometrics include portfolio choice with a large number of assets, derivative pricing
with time-varying parameters, etc. The Advisory Board would gladly consider and support even
more initiatives and interactions with practitioners of quantitative finance on these topics.
- Eric Renault, Member
8
Graduate Students: In Their Own Words
“My research topic was about the Social Accounting Matrix, which is an instrument to model the
„circular flow‟ of a market economy. The estimation of the matrix is a classic example of an illposed estimation problem: the number of estimated values exceeds the number of available data
points. The principle of maximum entropy provides a solid criteria for the estimation for such an
ill-posed problem. Furthermore, in comparison to other methods available, this application is
flexible and efficient. That is, various forms of information and an incomplete set of available
data may be incorporated into the estimation, while alternative methods are unable to. It is easier
said than done. The implementation involves an intensive data collection process, detailed
knowledge about SAM designs, and the use of non-linear optimization. The resources and
funding from the Info-metrics Institute have been very helpful, and particularly the support from
the senior researchers have been the most valuable for me to tackle those challenges.”
- Kentaro Murayama, Summer 2012 Fellow
“During this past summer, I studied the information-theoretic approach to estimating intervalvalued data. The aim of the research is to propose an interval estimation method in which
realistic distributions within the interval-valued data are preserved. In the future, I would like to
use my newly acquired knowledge from this research in order to apply these principles to highfrequency financial data and data from developing countries.
The Info-Metrics Institute's Summer Graduate Fellowship provided me a chance to not only
expose myself to the information and entropy literature, but also to sharpen my research skills
under the guidance of Prof. Amos Golan. I am extremely grateful to the Info-Metrics Institute for
giving me this great opportunity.”
- Tual Tuang, Summer 2012 Fellow
9
Thanks to our Sponsors
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 also like to
thank the Journal of Applied Econometrics for help in supporting the Information and
Econometrics of Networks conference in March. We also thank the Economics department for
partial support toward general Institute activities.
10
Institute Events 2011-2012
Institute participants gather for the beginning of the October 2011 Philosophy of Information
workshop
Luciano Floridi presents at the October 2011 Philosophy of Information workshop
11
Participants gather for a group photo during the November 2011 Shrinkage workshop
Eric Renault speaks during the Info-Metrics Theory and Shrinkage Estimation roundtable
12
David Wolpert presents at the March 2012 Networks conference
Roundtable participants at the March 2012 Networks workshop L-R: Nick Kiefer, Anil Bera,
M. Hashem Pesaran, Duncan Foley, Michael Stutzer, and Robin Lumsdaine
13
2011-2012 Advisory Board
Social Sciences
Duncan Foley (New School for Social Research and Santa Fe Inst.)
http://homepage.newschool.edu/~foleyd
John Geweke (U. Iowa and UTS, Sydney)
http://www.biz.uiowa.edu/faculty/jgeweke/
Alastair Hall (U. Manchester)
http://personalpages.manchester.ac.uk/staff/alastair.hall/
George Judge (Berkeley)
http://are.berkeley.edu/~judge/
Yuichi Kitamura (Yale)
http://cowles.econ.yale.edu/faculty/kitamura.htm
Robert Lerman, Chair (American U and Urban I)
http://american.edu/cas/faculty/rlerman.cfm
http://www.urban.org/about/RobertLerman.cfm
Robin Lumsdaine (American U)
http://www.american.edu/kogod/faculty/lumsdain.cfm
Esfandiar (Essie) Maasoumi (Emory)
http://userwww.service.emory.edu/~mboerck/
M. Hashem Pesaran (Cambridge)
http://www.econ.cam.ac.uk/faculty/pesaran/
Eric Renault (Brown)
http://news.brown.edu/new-faculty/2011-12/eric-renault
Peter M. Robinson (London School of Economics)
http://personal.lse.ac.uk/robinso1/
Richard Smith (Cambridge)
http://www.econ.cam.ac.uk/faculty/smith/
Halbert L. White, Jr. (1950-2012) (UC San Diego)
http://weber.ucsd.edu/~hwhite/
Arnold Zellner (1927-2010) (Chicago)
http://faculty.chicagobooth.edu/arnold.zellner/more/
14
Philosophy and the Natural Sciences
Pieter Adriaans (U. Amsterdam)
http://staff.science.uva.nl/~pietera/
Ariel Caticha (SUNY Albany)
http://www.albany.edu/physics/ariel_caticha.htm
Luciano Floridi (U of Oxford & U of Hertfordshire)
http://www.philosophyofinformation.net/About.html
Raphael D. Levine (The Hebrew U and UCLA)
http://www.fh.huji.ac.il/members/Levine/
http://faculty.chemistry.ucla.edu/institution/personnel?personnel_id=264627
15
2011-2012 Visiting Fellows
Fall 2011
Loran Grady Chollete
(December 6 - 9, 2011)
Info-Metrics Institute Visiting Fellow
UiS Business School, Norway
http://www.uis.no/chollete
Teddy Seidenfeld
(November 1 – 4, 2011)
Info-Metrics Institute Visiting Fellow
Departments of Philosophy and Statistics
Carnegie Mellon U
http://www.hss.cmu.edu/philosophy/faculty-seidenfeld.php
Spring 2012
Ariel Caticha
(May 21 – 25, 2012)
Info-Metrics Institute Visiting Senior Fellow
Professor of Physics SUNY–Albany
http://www.albany.edu/physics/ariel_caticha.htm
16
Affiliate Profiles
Ariel Caticha is professor at the Department of Physics, University at
Albany – SUNY. In recent years his research has focused on the connection
between physics and information.
One research goal has been to develop a general framework of entropic
inference that allows one to tackle in a unified manner the central issues –
the nature of information, how it ought to be processed, and the relation
between Bayesian and entropic methods.
The other goal has been to explore to which extent the laws of physics
might reflect the rules for processing information – the ambitious objective
is to derive the laws of physics from information theory. This approach has already yielded
multiple insights into statistical, quantum and classical mechanics.
Caticha devotes a considerable effort preparing lectures and tutorials on the subject of entropic
inference and information physics. He has received the SUNY Chancellor‟s Award for
Excellence in Teaching and the UAlbany Excellence in Teaching and Advising Award.
Michael Stutzer is professor of finance at the University of
Colorado in Boulder. He is an editorial board member for the
Journal of Futures Market, and served as executive director of the
Burridge Center for Securities Analysis and Valuation.
His theoretical work has focused on the application of
mathematical information theory, and its frequent interpretation
from the statistical theory of large deviations to well-known
problems in parameter estimation, derivative security pricing, and
portfolio choice.
His work is also well known among practitioners; he has won two awards from the CFA Institute
for work published in the Financial Analysts Journal. He is listed among the most frequently
read contributors to the Institutional Investor group of journals.
His current research projects include a large deviations approach to optimal futures hedging, and
a project exploring entropic methods for interbank financial fragility analysis.
17
Aman Ullah is distinguished professor of economics and chair at the
University of California, Riverside. He is an associate editor of
Econometric Reviews, Empirical Economics, American Journal of
Mathematical and Management Sciences, Macroeconomics and
Finance In Emerging Market Economies and other publications.
Aman advances the objectives of the Institute by sharing his extensive
knowledge in the disciplines of economics, econometrics, statistics,
and information theory. His wide-ranging areas of research include
nonparametric and semi-parametric methods, finite sample
econometrics, panel data econometrics, financial econometrics, and
econometrics of happiness.
He has been a Fellow of the Journal of Econometrics, American Association for Advancement of
Sciences (AAAS), National Academy of Sciences (India), and Royal Society of Statistics.
He recently co-edited the Handbook of Empirical Economics and Finance (Taylor and Francis,
2011).
18
Our Work – Advancing Interdisciplinary Info-Metrics Research
Following is a list of representative research and recent publications by some of the Institute‟s
affiliates.
Pieter Adriaans (University of Amsterdam)
Current Research
 Exploring the complexity of scientific data sets related to the COMMIT project, a large
Dutch cooperative effort in computer science
 Investigating various classes of complexity measures and developing software to analyze
RDF databases
 Developing a sound theoretical model of the concept of facticity
Other Updates
 Contributed a section on the philosophy of information for the Stanford Encyclopedia of
Philosophy (SEP).
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
Recent Publications
 An information theoretic approach to ecological inference in presence of spatial
dependence, Springer.
 Information theoretic methods in small domain estimation problems: The role of spatial
effects, Report - Blue-Ets Project (with E. Fernandez-Vazquez).
 Identification and estimation of club convergence models with spatial dependence,
International Journal of Urban and Regional Research.
 Non-linearities in economic growth and club convergence, Empirical Economics (with S.
Bertarelli).
Mehmet Caner (NC State University)
Current research:
 Adaptive lasso with heteroskedastic data
 Instrumental variable selection
 Recent publications:
 A Near Minimax Risk Bound: Adaptive Lasso with Heteroskedastic Data in Instrumental
Variable Selection (with M. Fan)
19
Marine Carrasco (University of Montreal)
Current Research
 Generalization of empirical likelihood estimation to handle a continuum or a countable
infinite number of moment conditions
 Regularization techniques of the covariance matrix and its application to portfolio
selection in finance
Forthcoming/Recent Publications
 Asymptotic normal inference in linear inverse problems, forthcoming in the Handbook of
Applied Nonparametric and Semiparametric Econometrics and Statistics, Oxford
University Press (with J. Florens and E. Renault).
 A regularization approach to the many instruments problem, Journal of Econometrics
(2012).
Luciano Floridi (University of Hertfordshire)
Current Research
 Understanding information quality standards and their challenges
 Ethics between online security and civil rights
Recent Publications
 The Philosophy of Information (volume one of the quadrilogy Principia Philosophiae
Informationis), Oxford University Press.
 Turing‟s three philosophical lessons and the philosophy of information, Royal Society‟s
Philosophical Transactions A.
 Semantic information and the network theory of account, Synthese.
Duncan Foley (New School for Social Research and Santa Fe Institute)
Current Research
 Ensemble complexity and its applications
Other Updates
 Teaching Graduate Econometrics at New School for Social Research in Spring 2013
A. Bowen Garrett (McKinsey & Co. and Urban Institute)
Current research:
 Economic and applied econometric modeling of health care and policy
 Using info-metric methods in micro-simulation modeling of health reforms
20
Ramo Gencay (Simon Fraser University)
Current Research
 Testing serial correlations with wavelets
 Testing jumps with wavelets
 Testing structural change with wavelets
 Counterparty risk and financial networks
Recent Publications
 Hierarchical information and the rate of information diffusion, Journal of Economic
Dynamics and Control (with Y. Xue
Amos Golan (American University)
Current Research
 The foundations of info-metrics
 Value of information
 Information dynamics
 The basics of modeling
Recent Publications
 Information Dynamics, Minds and Machines
 An Entropic Estimator for Linear Inverse Problems, Entropy (with H. Gzyl)
 On the Foundations and Philosophy of Info-Metrics, LNCS
Other Updates
 Info-Metrics Institute: Two Day Tutorial on Info-Metrics
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
Atsushi Inoue (NC State University)
Current Research
 Time series econometrics, including identification
macroeconomic models, forecasting, and factor models
21
and
misspecification
of
George Judge (University of California – Berkeley)
Forthcoming/Recent Publications
 An Information Theoretic Approach to Econometrics, Cambridge University Press (with
R. Mittelhammer).
 Not all empirical divergence are created equal, forthcoming in Proceedings of the
American Institute of Physics (with M. Grendar).
 A minimum means squared error semi parametric convex combining estimator,
forthcoming in Advances in Econometrics (with R. Mittelhammer).
 A risk superior semi parametric estimator for over identified models, forthcoming in
Advances in Econometrics (with R. Mittelhammer).
 A family of empirical likelihood estimators for the binary response model, Journal of
Econometrics (with R. Mittelhammer).
 On Extracting Probability Distribution Information from Time Series, Entropy (with A.
Kowalski, M. Martin, and A. Plastino).
Nick Kiefer (Cornell University)
Current Research
 Bayesian foundations for nonparametric methods
Robin Lumsdaine (American University)
Current research:
 Exploring the relationship between financial market perceptions and reality, and the role
of news and information in shaping those perceptions
 Testing conventional wisdoms in the inflation-linked bond market
 How survey design affects participant responses and subsequent inference
 The impact of the changing demographic landscape on the global financial markets
Essie Maasoumi (Emory University)
Forthcoming Publications
 Forthcoming paper with Levent Bulut in a significant Springer volume in honor of the
late Halbert White, Jr. (Demonstrates the application of metric entropy measures for
evaluation of exchange rate models and theories and similar debates about model
selection and prediction.)
Ilya Nemenman (Emory University)
Current Research
 Developing methods for information-theoretic statistical analysis of interaction networks,
22
focusing on cellular signaling as the application domain
Recent Publications
 Information transduction capacity of noisy biochemical signaling networks, Science (with
R. Cheong, A. Rhee, C.J. Wang, and I. Nemenman).
Jeffrey Racine (McMaster University)
Current Research
 Bandwidth selection for mixed data conditional and unconditional CDF and quantile
functions
 Nonparametric kernel estimation of mixed data copula functions
 Nonparametric instrumental variable estimation of functions and their derivatives
Forthcoming/Recent Publications
 Additive regression splines with irrelevant categorical and continuous regressors,
forthcoming in Statistica Sinica (with S. Ma).
 A smooth nonparametric conditional density test for categorical responses, forthcoming
in Econometric Theory (with C. Li).
 Smooth constrained frontier analysis: A festschrift in honour of Halbert L. White, Jr.,
forthcoming in Springer Verlag, X. Chen and N.E. Swanson Eds. (with C. Parmeter).
 Categorical semi parametric varying coefficient models, forthcoming in the Journal of
Applied Econometrics (with Q. Li and D. Ouyang).
 RStudio: A platform-independent IDE for R and Sweave, Journal of Applied
Econometrics.
 Jackknife model averaging, Journal of Econometrics (with B. Hansen).
Eric Renault (Brown University)
Current Research
 Volatility modeling with transaction data
 Stochastic volatility models for option pricing
 Inference with implied probabilities
 Multivariate volatility models
 Non-parametric inference in inverse problems
Recent Publications
 Affine fractional stochastic volatility models, Annals of Finance (with F. Comte and L.
Coutin).
 Efficient minimum distance estimation with multiple rates of convergence,
 Journal of Econometrics (with B. Antoine).
* Congratulations to Eric in his move to Brown as the C.V. Starr Professor of Economics.
23
Other Updates
 Co-editor of Econometric Theory
 Associate Editor of Econometrica
 President Elect of the Society for Financial Econometrics
Richard Smith (University of Cambridge)
Current Research
 General theory of estimation and inferential methods for econometric models specified
through moment conditions
 Development of reliable inferential methods in the moment condition context based on
the bootstrap and neglected heterogeneity in moment condition models
Forthcoming/Recent Publications
 GEL statistics under weak identification, forthcoming in the Journal of Econometrics
(with P. Guggenberger and J.J.S. Ramahlo).
 Discrete choice nonresponse, forthcoming in Review of Economic Studies (with E.J.
Ramahlo).
 Efficient aggregation of panel qualitative survey data: The construction of an indicator of
manufacturing output from the CBI survey in the United Kingdom, forthcoming in
Journal of Applied Econometrics (with J. Mitchell and M.R. Weale).
 GEL criteria for moment condition models, Econometric Theory.
David Wolpert (NASA Ames Research Center)
Forthcoming Publications
 Co-author of a book to be published by Springer of classic papers on information theory
and game theory.
Other Updates
 Held “Combining Information Theory and Game Theory,” a 2012 workshop cosponsored by the Santa Fe Institute and the New Mexico Consortium.
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
 Semi parametric estimation of shape-constrained functions
Recent Publications
 Maximum entropy approximations for asymptotic distributions of smooth functions of
24
sample means, Scandinavian Journal of Statistics (with S. Wang).
Victor Yakovenko (University of Maryland)
Forthcoming/Recent Publications
 Application of statistical mechanics to economics: Entropic origin of the probability
distributions of money, income, and energy consumption, (invited chapter in proceedings
of the symposium in honor of Duncan K. Foley on occasion of his 70th birthday at the
Department of Economics, New School for Social Research, New York, 20-21 April
2012) to be published by Routledge.
 Chapter 7: Statistical mechanics approach to the probability distribution of money, New
Approaches to Monetary Theory: Interdisciplinary Perspectives, Routledge.
 International studies in money and banking, Proceedings of the Workshop “Money Interdisciplinary Perspectives”, Department of Sociology, Free University of Berlin.
25
Summer 2012 Info-Metrics Summer Classes
Microeconometrics with focus on Panel Data and Discrete Choice: Theory and Practice
May 14-18, 2012
Instructor: William Greene (NYU)
5 day course
Day 1: May 14 – Basic statistics, linear regression, panel data for linear models
Session 1: Review
Statistics, quantiles, kernel density estimator
Linear Regression Model: Basic Methods estimation and inference
Bayesian vs. classical methods
Quantile regression
Bootstrapping
Session 2: Panel data,
Fixed and random effects linear models
Introduction to hierarchical models and random parameters
Session 3: Panel data methods:
Endogeneity, IV
Hausman and Taylor
Arellano/Bond/Bover dynamic models
Difference in Difference
Lab 1: Applications of linear regression and basic panel data methods
Day 2: May 15 - Nonlinear models, binary and ordered choice, count data
Session 1: Nonlinear Models
Estimation and inference in nonlinear models
Introduction to binary choice models
Session 2: Binary and ordered choice models
Specification and estimation
Inference
Analysis of nonlinear models
Ordered choice
Session 3: Models for count data.
Poisson and negative binomial
Two part models: zero inflation, hurdle models,
Lab 2: Applications of binary and ordered choice and count data models
Day 3: May 16 - Panel data and nonlinear models
Session 1 Fixed effects with nonlinear models
Probit model with fixed effects
Conditional estimation
Unconditional estimation
26
Incidental parameters problem
Estimation of models with fixed effects
Session 2 Random effects in nonlinear models
parameters and simulation
quadrature
cluster estimator
Session 3 Random parameters and hierarchical models
Lab 3 Applications of nonlinear models for panel data
Day 4: May 17 – Extensions of and applications of discrete choice and panel data models
Sessions 1-3 Discrete choice
Sample selection in nonlinear models
Sample selection in panel data models
Missing data methods
Latent class models
Lab 4: Discrete choice and panel data methods
Day 5: May 18 – Multinomial choice. Stated and revealed preference, mixed logit, etc.
Session 1: Multinomial choice models
Random utility
Random regret
Multinomial logit model
Multinomial probit
Session 2: Multinomial choice models
Nested logit
Heteroscedasticity and heterogeneity
Extensions of multinomial choice
Session 3: Mixed logit
Random parameters
Latent class
Generalized mixed logit
Stated choice applications
Lab 5: Multinomial choice models
27
Data Mining and Information: Theory and Practice
May 29-June 2, 2012
Instructors: Teddy Seidenfeld (Carnegie Mellon U), Richard Scheines (Carnegie Mellon
U), Kevin Zollman (Carnegie Mellon U)
Day 1 (Tuesday, May 29) Instructor: Teddy Seidenfeld.
Theme: Some information-based techniques in data-mining. Measures of statistical information
play a central role in statistical inference. We examine several information-based concepts in
statistical inference and their applications in data-mining. Topics to be covered include datareduction, prediction with sparse data, and techniques with undirected graphical models. We will
examine information-constraints operating in numerical algorithms, such as in the EM algorithm.
Day 2 (Wednesday, May 30) Instructor: Richard Scheines.
Theme: Graphical Causal Models and TETRAD. Tetrad is an integrated suite of free software for
specifying, estimating, and searching for graphical causal models. It contains over 20 algorithms
for searching for a variety of model classes, e.g., path analytic models, Bayes nets, factor
analytic and structural equation models, time series models general latent variable models of
conditional independence structures, and Markov blankets. In this workshop, we will briefly
introduce graphical causal models, show how to build, simulate data from, and estimate such
models in Tetrad, explain model search, teach how to use a variety of the search algorithms
in Tetrad on real and simulated data, and go through a few case studies in which Tetrad is
applied to real data. We will devote particular attention to comparing provably reliable causal
search algorithms to multiple regression, which is provably unreliable even on population
probabilities.
Day 3 (Thursday, May 31) Instructor: Teddy Seidenfeld.
Theme: Some information-based techniques in statistical decision-making. Statistical measures
of information also play an important role when gauging the value of new information for a
decision maker. What is the value of new information? We will examine the value of information
both for asymptotic and short-run decision problems, involving cases with both single and
multiple decision makers. Topics to be covered include the role of information in experimental
design, and in multi-agent problems, such as in games, where there are issues of strategic
interaction.
Day 4 (Friday, June 1) Instructor: Kevin Zollman.
Theme: Strategic emergence of information transfer when there is information in a system there
are a number of important issues about how that information comes to be transmitted by
economic actors. In the case of simple two-agent interactions, information might be strategically
released, strategically modified, or withheld altogether. There are a number of (game theoretic)
analyses that outline the conditions under which one should expect information to be transmitted.
As one moves from a two-agent situation to a multi-agent situation, things become even more
complex. Even when everyone desires that information be transmitted, forming optimal
institutions for the transmission of that information can be difficult. This day will focus on both
the issue of strategic information transmission and the strategic formation of institutions for the
transmission of information.
28
Day 5 (half-day: Saturday morning, June 2) Instructor: Teddy Seidenfeld.
Theme: Information and some non-parametric Graphical and Bayesian methods.
We illustrate inference methods for large data sets using a combination of ideas
developed in previous lectures, including variational and hierarchical models.
29
A Special Two-Day Tutorial on Info-Metrics
July 9-10, 2012
Instructor: Amos Golan (American U)
Background
All learning, information gathering and information processing, is based on limited knowledge,
both a priori and data, from which a larger “truth” must be inferred. To learn about the true state
of the world that generated the observed data, we use statistical models that represent these
outcomes as functions of unobserved structural parameters, parameters of priors and other
sampling distributions, as well as complete probability distributions. Since we will never know
the true state of the world, we generally focus, in statistical sciences, on recovering information
about the complete probability distribution, which represents the ultimate truth in our model.
Therefore, all estimation and inference problems are translations of limited information about the
probability density function toward a greater knowledge of that probability density function.
Information theory quantities, concepts, and methods provide a unified set of tools for organizing
this learning process. They provide a discipline that at once exposes more clearly what the
existing methods do, and how we might better accomplish the main goal of scientific learning. In
this his course we will study the basic quantities of information theory and their relationships to
data analysis and information processing, and then use these quantities to study and develop
improved methods of estimation and data processing that compose the class of entropy and
information-theoretic methods. Within the above class of Information-Theoretic models, this
course concentrates on methods that use conditional and unconditional stochastic moments.
The Complete Course
This course concentrates on the statistical and econometric analysis of problems and data in the
social sciences. The methods used for analyzing these problems are within the sub-field of
Information-Theoretic estimation methods. The simple problem of estimating more unknown 2
parameters than known observations is a basic example for a common problem within IT. To
solve these types of problems, one resorts to the quantity known as “entropy”. That quantity is
related to the concept of information. In the first part of the course, we will study the basic
concepts of entropy and information. We will use these quantities to solve under-determined
problems. In the next part of the course, we will extend this approach for analyzing all types of
data. We will discuss both linear models and nonlinear models, including discrete choice and
censored models. The course will consist of five days of class lectures and computer laboratories
sessions and tutorials. We will work with real and artificial data.
The Two-Day tutorial will cover only part of the above material.
This two-day tutorial involves some computing. For those who are used to write their own
computer codes, these computing can be done by using different software, such as Matlab,
Gauss, GAMS, etc. For those who wish to use common econometric software, the models we
discuss in this class, can be used within the main econometric software packages: SAS,
LIMDEP. In this tutorial, we will use GAMS, SAS and LIMDEP.
30
Topic Outline
1. Background, motivation and philosophy
Why Info-Metrics?
Brief History
The foundations of information theory
What is Entropy
The axiomatic and combinatorial derivations of the entropy measure
The basic problem
The classical Maximum Entropy (ME) principle and formulation
The dual (concentrated - unconstrained) formulation
Prior information
Basic diagnostics and test-statistics
Comparison with the standard Maximum Likelihood (ML) and other estimation methods.
2. Derivation of the basic Generalized ME (GME) method – A simple example
Consider recovering the unknown coefficients of a matrix (matrix balancing) from aggregated
data (e.g., estimating the coefficients of Input-Output Table, a Social Accounting Matrix, or first
order Markov process). A complete comparison with the traditional methods (e.g., ML) will be
developed. Extensions that allow incorporating more variables (e.g., macro/policy) and
accommodating for noisy data will be discussed in great detail.
The basic pure problem
The Maximum Entropy solution
The Concentrated problem
The ML Solution
ME – ML
The real, noisy problem3
The GME solution
Inference and diagnostics
3. The traditional linear statistical regression model
The basic set-up of the problem
Basic derivation of the GME
Primal vs. dual formulations
Diagnostics
Derivations and A comparisons with other information-theoretic methods
Depending on time availability we may discuss one/more of the topics below – but not all
4. Extensions of the linear model (autocorrelations, heteroskedasticity, etc)
5. Discussion of other IT Estimators
6. Discussion of theoretical and empirical applications
7. Additional topics
31
Discrete choice models (ordered/unordered);
Set of equations and simultaneous equations;
8. Summary and discussion of possible future directions
32
2011-2012 Institute Seminars
Fall 2011
Victor Yakovenko (University of Maryland)
"Entropy maximization and distributions of money, income, and energy consumption in a market
economy"
Robin Lumsdaine (American U)
"How Survey Design Affects Inference Regarding Health Perceptions and Outcomes"
Teddy Seidenfeld (Carnegie Mellon U)
"Bayesian Consensus"
Loran Grady Chollete (UiS Business School, Norway)
"The Market Premium for Dynamic Tail Risk"
Spring 2012
Richard Davis (Columbia U)
“Detection of Structural Breaks and Outliers in Time Series”
Kees Bouwman (Erasmus University)
"Stock Market Liquidity and Bond Risk Premia"
Sherman Robinson (IFPRI)
"Applications of Maximum Entropy Econometrics to Estimation Problems in Developing
Countries"
Herman K. van Dijk (Erasmus University Rotterdam and Econometrics Department VU
University Amsterdam)
"Is Simulation based Bayes Procedures for Three 21-st Century Key Research Issues"
33
2011-2012 Workshops & Conferences
October 2011 Workshop: Philosophy of Information
Workshop Objectives
The overall objective of this workshop is to study some of the open questions within philosophy
of information.
Interest in the philosophy and meaning of information goes back half a century but has rapidly
increased recently with many new directions of research into the meaning, quantification and
measures of information and complexity as well as a vast range of applications across the
scientific spectrum.
In this conference we will focus on just one aspect of the philosophy of information: the different
techniques to measure information and to identify meaningful information. Various definitions
(of meaningful information) have been proposed in the recent past: sophistication, effective
complexity, computational depth, self-dissimilarity and facticity. At a first glance these
definitions seem to cover various aspects of the same domain. If, for example, one defines
meaningful information in terms of a balance between ad hoc and structural information in a
dataset then this definition automatically has consequences for self dissimilarity, the effective
complexity, the facticity and the computational depth of the set. A natural question is what are
the exact relations between these various definitions, and whether it is possible to reduce these
concepts into one (or more) central concept/s. Another interesting question is which kind of
processes in nature and society creates meaningful information: evolution, game playing, stock
exchange, etc.
This one day conference will address these basic questions and will explore recent advances in
the philosophy of information and its potential applications.
Conference Topics
Measuring information
How to define and quantify meaningful information
What natural (or behavioral) processes create meaningful information?
Program Committee
Pieter Adriaans (U. Amsterdam) – Co-Chair
Duncan Foley (New School for Social Research and Santa Fe Inst.)
Amos Golan (American U.) – Co-Chair
Esfandiar (Essie) Maasoumi (Emory)
34
Confirmed Speakers
Pieter Adriaans (U. Amsterdam)
Luis Antunes (Porto U., Portugal)
Ariel Caticha (SUNY Albany)
J. Michael Dunn (Indiana U. Bloomington)
Luciano Floridi (Oxford U. and U. Hertfordshire)
Duncan Foley (New School for Social Research and Santa Fe Institute)
David Oliver
Workshop Program
8:00-9:00 AM Registration
9:00-9:10 AM Welcoming Remarks
Peter Starr (American U.)
9:10-9:20 AM Info-Metrics and Workshop's Objectives
Amos Golan (American U)
9:25-10:10 AM Session I
Chair: Nathan Harshman (American U.)
"The Quantum Origin of New Information"
David Oliver (Meadows of Dan, Virginia, USA)
10:15-11:00 AM Session II
Chair: Anthony F. Beavers (U. of Evansville)
"Information Closure and the Sceptical Objection"
Luciano Floridi (Oxford U. and U. Hertfordshire)
11:00-11:15 AM Coffee break
11:15-12:00 PM Session III
Chair: Robin Lumsdaine (American U.)
“Contradictory Information: Better than Nothing”
J. Michael Dunn (Indiana U. Bloomington)
12:00-1:00 PM Lunch
1:00-2:30 PM - Session IV
Chair: Eric Renault (Brown)
"Facticity and the quest for meaningful information"
Pieter Adriaans (U. Amsterdam)
"Notes on facticity and effective complexity"
35
Duncan Foley (New School for Social Research and Santa Fe Inst.) and David Oliver (Meadows
of Dan, Virginia, USA)
2:30-3:15 PM Session V
Chair: Robert Lerman (American U., Urban Institute)
“Information and Entropy in a Bayesian Framework”
Ariel Caticha (SUNY Albany)
3:15-3:45 PM Coffee break
3:45-4:30 PM Session VI
Chair: Richard Scheines (Carnegie Mellon)
"Useful Information"
Luis Antunes (Porto U., Portugal)
4:30-4:45 PM Coffee break
4:50-6:15 PM – Roundtable: Info-Metrics: Philosophy of Information
Moderator: Pieter Adriaans (U. Amsterdam)
Anthony F. Beavers (U. of Evansville)
Luciano Floridi (Oxford U. and U. Hertfordshire)
Aidan Lyon (UMD)
Brian Ziebart (Carnegie Mellon U)
6:15 -7:30 PM Reception
October 2011 Workshop papers and additional materials can be found on the workshop
webpage: http://www.american.edu/cas/economics/info-metrics/workshop/workshop-2011fall.cfm
36
November 2011 Workshop: Information Theory and Shrinkage Estimation
The conference is jointly hosted by the Info-Metrics Institute and North Carolina State
University. The co-chairs are Mehmet Caner (NCSU), Amos Golan (AU), and Peter Phillips
(Yale).
Workshop Objectives
Interest in shrinkage estimators goes back half a century but has rapidly increased recently with
many new directions of research that cover a vast range of applications in different disciplines.
Ongoing research on information-theoretic estimation and inference methods is similarly interdisciplinary, involving information theory, engineering, mathematical statistics, econometrics
and the natural sciences.
Modern shrinkage estimators like adaptive Lasso and Bridge methods are penalized estimators
with various penalty designs. These methods assist in model selection, oracle efficient estimation
and provide substantial computational advantages. In these respects they may be viewed as
methods that utilize information better than existing techniques. They have been found to be
particularly helpful in large dimensional systems where shrinkage can deliver sparse estimation,
sparsisity in the limit, and improved finite sample performance. The range of applications is vast
and includes microeconometric models of labor supply, time series econometric applications to
economic growth, aggregate cointegrated systems, and market wide financial volatility, social
interaction network estimation, and gene selection in biology.
This one day conference will address these various themes, the inter-connections between
shrinkage estimation methodology and info-metrics, and explore recent advances in shrinkage
methods and applications.
Workshop Topics
Shrinkage Estimators and Info-Metrics
Regularization Methods and Info-Metrics
Model Discovery and Info-Metrics
Program Committee
Mehmet Caner (NC State U) – Co-Chair
Amos Golan (American U) – Co-Chair
George Judge (UC Berkeley)
Robin Lumsdaine (American U)
Peter Phillips (Yale) – Co-Chair
Eric Renault (Brown)
37
Keynotes Speakers
Andrew Barron (Yale)
Victor Chernozhukov (MIT)
Harrison Zhou (Yale)
Other Confirmed Speakers
Mehmet Caner (NCSU)
Marine Carrasco (Montreal)
Bruce Hansen (Wisconsin)
George Judge (UC Berkeley)
Eric Renault (Brown)
Song Song (UC Berkeley)
Pascale Valery (HEC-Montreal)
Liao Zhipeng (UCLA)
Workshop Program
8:00-9:00 AM Registration
9:00-9:10 AM Welcoming Remarks
Amos Golan (American U)
9:10-9:25 AM Conference Opener and Objectives
Amos Golan (American U)
Peter Phillips (Yale)
9:25-9:55 AM Invited Talk: Victor Chernozhukov (MIT)
“Least Squares After Model Selection”
Chair: Teddy Seidenfeld (Carnegie Mellon)
9:55-10:05 AM Discussion Harrison Zhou (Yale)
10:10-11:00 AM Session I: Shrinkage Estimators in Econometrics
Chair: Eric Renault (Brown)
1. Liao Zhipeng (UCLA) and Peter C.B. Phillips (Yale)
“Automated Estimation of Vector Error Correction Models”
Discussion: Mehmet Caner (NCSU)
38
2. Mehmet Caner (NCSU) and Michael Fan (NCSU)
“A near minimax risk bound: adaptive lasso with heteroskedastic data in instrumental variable
selection”
Discussion: Liao Zhipeng (UCLA)
11:00-11:15 AM Coffee break
11:15 AM-12:05 PM Session II: Regularization
Chair: Essie Maasoumi (Emory U.)
1. Marine Carrasco (Montreal) and Neree Noumon (Montreal)
“Optimal Portfolio Selection using Regularization”
Discussion: Pascale Valery (Montreal)
2. Jean Marie Dufour (HEC Montreal) and Pascale Valery (HEC-Montreal)
“Wald type tests when rank conditions fail: a smooth regularization approach”
Discussion: Marine Carrasco (Montreal)
12:10-1:00 PM Lunch
1:00-2:05 PM - Session III: High Dimensional Covariances
Chair: Aman Ullah (UC Riverside)
1. Invited Talk: Harrison Zhou (Yale)
“High-Dimensional Covariance Structure Estimation”
Discussion: Victor Chernozhukov (MIT)
2. Song Song (UC Berkeley)
“Dynamic large spatial covariance matrix estimation in application to semiparametric model
construction via variable clustering; the SCE approach”
Discussion: Andrew Barron (Yale)
2:10-3:30 PM Session IV: New Estimators Via Information Theory
Chair: Marine Carrasco (Montreal)
1. Bruce Hansen (Wisconsin) “Generalized Shrinkage Estimators”
Discussion: Essie Maasoumi (Emory U.)
2. Jonathan B. Hill (UNC-CH) and Eric Renault (Brown)
“GMM with Tail Trimming”
Discussion: Amos Golan (American U.)
3:30-3:45 PM Coffee break
39
3:45-4:15 PM - Invited Talk: Andrew Barron (Yale)
"Statistical Risk Reduction by Combining or Selecting Models"
Chair: Robin Lumsdaine (American U)
4:15-4:25 PM Discussion: Song Song (UC Berkeley)
4:25-5:00 PM - Short Presentation/Poster Session
Chair: Peter Phillips (Yale)
a. Peter Exterkate (CREATES, Aarhus University)
"Nonlinear forecasting with many predictors using kernel ridge regression"
b. Enrique Pinzon-Garcia (Wisconsin)
“Linear Regression with a large number of weak instruments using post l-1 penalized estimator”
c. Chu An Liu (Wisconsin)
"A Plug-In Averaging Estimator for Regressions with Heteroskedastic Errors"
d. Shi Zhentao (Yale)
“Estimation of High Dimensional Structural Model”
e. Jerome Krief (LSU)
“Instrumental Nonparametric Estimation Under Conditional Moment Restrictions: A Nonlinear
Tikhonov Approach”
f. Anders Bredahl Kock (CREATES and Aarhus University)
"Oracle Efficient Variable Selection in Random and Fixed Effects Panel Data Models"
5:00-5:15 PM - Coffee break
5:15-6:15 PM - Round Table – Information Theory and Shrinkage Estimation
Moderator: Mehmet Caner (NCSU)
Bill Greene (NYU)
Nick Kiefer (Cornell)
Robin Lumsdaine (American U.)
Essie Maasoumi (Emory U.)
Peter Phillips (Yale)
Eric Renault (Brown)
Aman Ullah (UC Riverside)
6:15 -7:30 PM – Reception
November 2011 Workshop papers and additional materials can be found on the workshop
webpage: http://www.american.edu/cas/economics/info-metrics/workshop/workshop-2011november.cfm
40
March 2012 Workshop: Information and Econometrics of Networks
Workshop Objectives
The two-day conference is supported by the Journal of Applied Econometrics and will be
organized jointly by Professors Amos Golan of American University and Essie Maasoumi of
Emory University. Social and economic networks are everywhere: from Facebook to the more
complex global financial network or to networks connecting economic agents or to other
complex and dynamic economic networks. The study of these networks is crucial for both
academics and policy makers and presents a host of new theoretical and econometric challenges.
This conference will concentrate on studying the information and econometrics of networks. The
conference organizers encourage submissions of papers on any topic within this overall
theme with a particular emphasis on the list below.
Workshop Topics
Info-Metrics and Networks
Spatial Econometrics
Information in Networks
Factor Dependence
Empirical Analysis of Social and Economic/Financial Networks
Financial Networks and Risk
Complex Networks
Program Committee
Anil Bera (U Illinois, Urbana-Champaign)
Amos Golan (American U.), Co-Chair
Robin Lumsdaine (American U.)
Esfandiar (Essie) Maasoumi (Emory), Co-Chair
Michael J. Stutzer (U. Colorado-Boulder)
Confirmed Speakers
Marco Avellaneda (NYU)
Larry Blume (Cornell)
Ethan Cohen-Cole (U. Maryland)
Rama Cont (Columbia)
Steven N. Durlauf (U. Wisconsin, Madison)
Craig Friedman (S&P)
Michael Gordy (Federal Reserve Board)
Jiawei Han (U. Illinois at Urban-Champaign)
Peter Robinson (London School of Economics)
Dan Rockmore (Dartmouth)
Michael J. Stutzer (U. Colorado-Boulder)
David Wolpert (NASA)
41
Program
Friday, March 30
8:30-9:30 Registration
9:30-9:35 AM Welcoming Remarks
Peter Starr (Dean of College of Arts & Sciences, American U)
9:35-10:00 AM Info-Metrics and Workshop's Objectives
Amos Golan (American U)
10:00-12:00 PM SESSION I: Info-Metrics, Systemic Risk and Finance - I
Session Organizer – Michael J. Stutzer
Chair: Duncan Foley (New School for Social Research and Santa Fe Inst.)
1. Entropy in Financial Network Contagion Research Michael J. Stutzer (U. Colorado)
2. Systemic Risk and Network Formation in the Interbank Market Ethan Cohen-Cole (U.
Maryland)
3. Contagion and systemic risk in interbank networks Rama Cont (Columbia)
4. Trading Networks Celso Brunetti (Federal Reserve Board)
12:10-1:00 PM SESSION II: Network Groups and Econometrics
Chair: Robin Lumsdaine (American U)
Exponent of Cross-sectional Dependence: Estimation and Inference Natalia Bailey (Cambridge),
George Kapetanios (U. London), M. Hashem Pesaran (Cambridge and U. of Southern
California)
1:00-2:00 PM Lunch
2:00-3:15 PM SESSION III: Economic and Decision Networks
Chair: Robert Lerman (American U and Urban Institute)
1. Linear Social Network Models, Steven N. Durlauf (U. Wisconsin), Lawrence Blume,
William Brock, and Rajshri Jayaraman
2. Main Street vs. Wall Street: A Comparison of Views Robin Lumsdaine (American U) and
Rogier J.D. Potter van Loon (Erasmus U)
3:15-3:30 PM Coffee Break
42
3:30-4:30 PM Session IV: Social Networks
Chair: Anil Bera (U. Illinois)
1. Identification and Estimation of Social Interactions through Variation in Equilibrium Influence
Mikko Packalen (University of Waterloo)
2. A Social Interactions Model with Endogenous Friendship Formation and Selectivity ChihSheng Hsieh and Lung-Fei Lee (Ohio State University)
3. A Structural Model of Segregation in Social Networks Angelo Mele (Johns Hopkins)
4. The Role of Reciprocation in Social Network Formation, with an Application to
Blogging,Alexia Gaudeul (Friedrich Schiller University) and Caterina Giannetti (University of
London)
4:30-4:40 PM Coffee Break
4:40-6:10 PM Session V: Complex and Information Networks
Chair: Aman Ullah (UC Riverside)
1. Complex Systems and Economics -- A Network Theoretic Approach Daniel N. Rockmore
(Dartmouth)
2. Unawareness, Information Theory, and Multiagent Influence Diagrams David Wolpert (Santa
Fe Inst. and Information Sciences Group, Los Alamos National Laboratory)
6:15-7:45 PM Reception
Saturday, March 31
9:00-10:00 AM SESSION VI: Info-Metrics, Systemic Risk and Finance - II
Session Organizer – Michael J. Stutzer
Chair: Michael J. Stutzer (U. Colorado)
1. Finding diverse and manageable sets of plausible yet severe financial system scenarios Craig
Friedman (S&P)
2. Credit Default Swaps and Systemic Risk Andreea Minca (Cornell)
3. Strategic Behavior and Endogenous Risk of Contagion in a Financial Network: A Network
Formation Game Pallavi Baral (Indiana University)
10:05-10:50 AM SESSION VII: Financial and Social Networks
Chair: Dennis Glennon (OCC)
1. Market Microstructure Implications of Financial Networks Naomi Boyd (West Virginia
University), Abel Rodriguez (University of California Santa Cruz), Brenda Betancourt
(University of California-Santa Cruz)
2. Economic Links and Counterparty Risk: The Network Determinants of Credit Spreads Daniel
Signori (Simon Fraser University) and Ramazan Gencay (Simon Fraser University)
3. Point Process Modeling for Directed Interaction Networks Patrick J. Wolfe joint work with
Patrick Perry, NYU (University College London, Harvard University)
43
10:50-11:10 AM Coffee Break
11:10-12:00 PM SESSION VIII: Complex Information Networks - II
Chair: Nathan Harshman (American U)
1. Mining Heterogeneous Information Networks: Do We Need New Metrics? Jiawei Han, (U.
Illinois at Urbana-Champaign)
12:00-1:00 PM - Lunch
1:00-1:45 PM SESSION IX: Economic and Financial Networks - Applications
Chair: Simon Sheng (American U.)
1. Nonparametric Analysis of Two-Sided Markets Senay Sokullu (University of Bristol)
2. Endogenous Response to the Interbank Loan 'Network Tax': Stability and Efficiency José
Pedro Fique (Universidade do Porto, Portugal)
2. A Model of Endogenous Extreme Events Loran Grady Chollete (University of Stavanger)
1:45-3:30 PM Round Table – Info-Metrics: Information Processing across the Sciences
Moderator: Robin Lumsdaine (American U)
Anil Bera (U Illinois)
Duncan K. Foley (New School for Social Research and Santa Fe Inst.)
Nicholas Kiefer (Cornell U)
M. Hashem Pesaran (Cambridge and U. of Southern California)
Michael Stutzer (U Colorado)
March 2012 Workshop papers and additional materials can be found on the workshop webpage:
http://www.american.edu/cas/economics/info-metrics/workshop/workshop-2012-spring.cfm
44
Forthcoming Workshops & Conferences (2011-2012)
Fall 2012 Workshop: November 17, 2012
Title: Info-Metrics and Nonparametric Inference
Location: University of California Riverside
Sponsors
US Comptroller of the Currency
UC Riverside, Department of Economics
Conference Objectives
The one day conference is organized jointly by the Info-Metrics Institute, American University
and the Department of Economics of University of California, Riverside.
Interest in nonparametric estimation and inference goes back half a century but has rapidly
increased recently (especially with recent advances in computing power) with many new
directions of research that cover a vast range of applications in different disciplines. Ongoing
research on information-theoretic estimation and inference methods is similarly interdisciplinary, involving information theory, engineering, mathematical statistics, econometrics
and the natural sciences.
This one day conference will explore recent advances in the area of nonparametric estimation
and inference and in info-metrics, which may help current and future research combining
nonparametric procedures with information-theoretic methods.
The conference organizers encourage submissions of papers on any topic within this overall
theme with a particular emphasis on the list below.
Conference Topics
Nonparametric procedures and info-metrics methods
Nonparametric and Information-Theoretic Estimation and Inference Methods
Density and Regression Estimation (Nonparametric and Information Methods)
Kernel estimation and info-metrics
Mixed data estimation
Quantile estimation
Nonparametric and Information time series analysis
Nonparametric and Information based panel data models
45
Program Committee
Amos Golan (Info-Metrics, American U) - Co-Chair
Aman Ullah (UC Riverside and Info-Metrics) - Co-Chair
Robin Lumsdaine (Info-Metrics, American U) - Co-Chair
Tae Hwy Lee (UC Riverside) - Co-Chair
Jeff Racine (McMaster University)
Nick Kiefer (Cornell)
Confirmed Speakers
John Fisher (MIT)
Nicholas (Nick) Kiefer (Cornell U)
Essie Maasoumi (Emory U)
Rosa L. Matzkin (UCLA)
Whitney Newey (MIT)
Jeffrey Racine (McMaster U)
Eric Renault (Brown U)
Peter Robinson (London School of Economics)
November 2012 Workshop webpage: http://www.american.edu/cas/economics/infometrics/workshop/workshop-2012-november.cfm
46
Spring 2013 Workshop: April 26, 2013
Title: Philosophy of Information, The Value of Information
Location: American University
Sponsors
US Comptroller of the Currency
AU College of Arts & Sciences
Workshop Objectives
The overall objective of this workshop is to study some of the open questions within philosophy
of information with an emphasis on the study of the value of information and the philosophy of
information processing.
Interest in the philosophy and meaning of information goes back half a century but has rapidly
increased recently with many new directions of research into the meaning, quantification and
measures of information and complexity as well as a vast range of applications across the
scientific spectrum.
In our first workshop on the topic (October, 2011) we focused on one aspect of the philosophy of
information: the different techniques to measure information and to identify meaningful
information.
Building on our earlier workshop and recent development, in this workshop we focus on a
number of related aspects of the philosophy of information and the philosophy of info-metrics.
These topics include the value of information, quantifying information, processing
complementary and contradicting information and the inter-relationship between information,
computation and complexity.
This one day workshop will address these basic questions and will explore recent advances in the
philosophy of information and its potential applications.
Workshop Topics
Value of information
Quantifying information
Measuring information
Processing information
How to define and quantify meaningful information
Program Committee
Pieter Adriaans (U. Amsterdam)
Ariel Caticha (SUNY Albany)
47
Luciano Floridi (Hertfordshire and Oxford), Co-Chair
Amos Golan (American U.), Co-Chair
Confirmed Speakers
Pieter Adriaans (U. Amsterdam)
Min Chen (Oxford)
J. Michael Dunn (Indiana University, Bloomington)
Luciano Floridi (U of Oxford & U of Hertfordshire)
Oliver Gossner (Paris School of Economics and London School of Economics)
James Moor (Dartmouth)
April 2013 Workshop webpage: http://www.american.edu/cas/economics/infometrics/workshop/workshop-2013-spring.cfm
48
Support the Institute!
The Info-Metrics Institute is happy to receive donations toward its different activities.
Contributions to the Info-Metrics Institute are tax deductible, subject to federal and state
guidelines.
With these resources, we hope to establish more long-term 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 Aisha Khan at
info-metrics@american.edu or 202-885-3770.
49
For More Information
Visit our website: www.american.edu/info-metrics
Sign up for our newsletter: send an email to info-metrics@american.edu
Contact us!
info-metrics@american.edu (Institute and webmaster)
Professor Amos Golan, Director of Info-Metrics Institute
Email: agolan@american.edu
Aisha Khan, Institute Administrator
Email: malik@american.edu
50