DIMACS Workshop on Markets as Predictive Devices (Information Markets)

Linda Casals lindac at dimacs.rutgers.edu
Wed Nov 3 11:15:50 EST 2004


*************************************************
     
DIMACS Workshop on Markets as Predictive Devices (Information Markets)
          
     February 2-4, 2005
     DIMACS Center, Rutgers University, Piscataway, NJ

Organizers: 
    
     Robin Hanson, George Mason University, rhanson at gmu.edu 
     John Ledyard, California Institute of Technology,
       jledyard at hss.caltech.edu 
     David Pennock, Overture Services, David.Pennock at overture.com 
     
Presented under the auspices of the Special Focus on Computation and
the Socio-Economic Sciences.

    ************************************************ 

For decades, economists have studied an astonishing "side effect''
of financial and wagering markets: their ability to serve as highly
accurate forecasting devices. This workshop aims to explore the use of
markets as a substitute for, or complement to, more traditional
forecasting tools. We will examine how information flows from traders
to the market and back again, how market mechanisms process
information, how market prices communicate information and forecasts,
and what mechanisms best foster accurate and statistically-testable
predictions. The workshop will bring together researchers and
practitioners from a variety of relevant fields, including economics,
finance, computer science, and statistics, in both academia and
industry, to discuss the state of the art today, and the challenges
and prospects for tomorrow.

A market designed from the outset for information gathering and
forecasting is called an information market. Information markets can
be used to elicit a collective estimate of the expected value or
probability of a random variable, reflecting information dispersed
across an entire population of traders. The market prediction is not
usually an average or median of individual opinions, but is a complex
summarization reflecting the game-theoretic interplay of traders as
they obtain and leverage information, and as they react to the actions
of others obtaining and leveraging their own information, etc. In the
best case scenario, the market price reflects a forecast that is a
perfect Bayesian integration of all the information spread across all
of the traders, properly accounting even for redundancy. This is the
equilibrium scenario called rational expectations in the economics
literature, and is the assumption underlying the strong form of the
efficient markets hypothesis in finance.

The degree to which market forecasts approach optimality in
practice, or at least surpass other known methods of forecasting, is
remarkable. Supporting evidence can be found in empirical studies of
options markets, commodity futures markets, political stock markets,
sports betting markets, horse racing markets, market games, laboratory
investigations of experimental markets, and field tests. In nearly all
these cases, to the extent that the financial instruments or bets are
tied to real-world events, market prices reveal a reliable forecast
about the likely unfolding of those events, often beating expert
opinions or polls.

Despite a growing experimental literature, many questions remain
regarding how best to design, deploy, analyze, and understand
information markets, including both technical challenges (e.g.,
designing combinatorial exchanges and social challenges (e.g.,
overcoming legal and ethical concerns). The search for answers will
benefit from input from economists (including specialists in mechanism
design, experimental economics, financial markets, wagering markets,
and rational expectations theory), statisticians and decision
theorists (including experts in forecasting, belief aggregation, group
decision making, Bayesian updating, and opinion polling), and computer
scientists (including experts in combinatorial exchanges, distributed
computing, information theory, and mixing worst-case and Bayesian
analysis). This workshop will seek to bring together a variety of
experts representing these fields, to engage in a dialog describing
current and future research directions to facilitate the design,
refinement, and proliferation of markets as predictive devices.

As part of the workshop, one or more tutorials are planned for the
benefit of students and other newcomers to the field; little or no
background knowledge will be assumed.

**************************************************************
Call for Participation:

This workshop will include talks on information markets (a.k.a,
prediction markets, event markets, or idea futures) by a number of
distinguished invited speakers. Speakers will cover a range of topics
including mechanism design, experiments, analysis, policy, and
industry experience. Speakers will include representatives from
academia, industry, and government. The workshop will feature research
talks, opinions, reports of industry experience, and discussion of
government policy from the perspective of a number of fields,
including economics, business, finance, computer science,
gambling/gaming, and policy. See the workshop program for more
details.

The workshop will feature a tutorial session on Wednesday afternoon
(Feb. 2, 2005) to help those new to the field get up to speed. The
workshop will include a panel discussion on the Policy Analysis Market
(a.k.a., "Terror Futures") and a "rump" session where anyone who
requests time can have the floor for five minutes to speak on any
relevant topic. To participate in the rump session, please email David
Pennock at pennockd at yahoo-inc.com.

**************************************************************
Registration Fees:

(Pre-registration deadline: January 26, 2005)

Please see website for additional registration information.

*********************************************************************
Information on participation, registration, accomodations, and travel 
can be found at:

http://dimacs.rutgers.edu/Workshops/Markets/

   **PLEASE BE SURE TO PRE-REGISTER EARLY**

*******************************************************************


---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com



More information about the cryptography mailing list