Distributed prediction markets modeled by weighted Bayesian graphical games

Janyl Jumadinova, Prithviraj Dasgupta

Research output: Contribution to journalArticle

Abstract

We consider a novel, yet practical setting of prediction markets called distributed prediction markets, where the aggregated price of a security of an event in one prediction market is affected dynamically by the prices of securities of similar events in other, simultaneously running prediction markets. We focus on the problem of decision making facing a market maker to determine the price of a security within such a setting. We propose a formal framework based on graphical games called a weighted Bayesian graphical game (WBGG) to model the distributed prediction market setting and to capture the local interactions between multiple market makers. We then describe a distributed message passing algorithm based on NashProp algorithm to calculate the Bayes-Nash equilibrium in a WBGG. We provide analytical results including convergence and incentivizing truthful revelation among market makers. Our experimental results show that market makers that consider the influence of other market makers in a distributed prediction market setting while using our proposed WBGG-based algorithm obtain higher utilities and set prices more accurately in comparison to market makers using a greedy strategy to set prices or those that do not consider the influence of other market makers. We also observe that extreme sizes of the neighborhood of a market maker have an adverse impact on its utilities.

Original languageEnglish (US)
Pages (from-to)84-98
Number of pages15
JournalLecture Notes in Business Information Processing
Volume187
DOIs
StatePublished - Jan 1 2014

Fingerprint

Game
Prediction
Message passing
Graphics
Market
Prediction markets
Market makers
Decision making
Message-passing Algorithms
Local Interaction
Bayes
Distributed Algorithms
Nash Equilibrium
Convergence Results
Extremes
Decision Making
Calculate
Experimental Results

Keywords

  • Bayes-Nash
  • Distributed
  • Graphical games
  • Prediction market

ASJC Scopus subject areas

  • Management Information Systems
  • Control and Systems Engineering
  • Business and International Management
  • Information Systems
  • Modeling and Simulation
  • Information Systems and Management

Cite this

Distributed prediction markets modeled by weighted Bayesian graphical games. / Jumadinova, Janyl; Dasgupta, Prithviraj.

In: Lecture Notes in Business Information Processing, Vol. 187, 01.01.2014, p. 84-98.

Research output: Contribution to journalArticle

@article{02f389746ba4405d938d3147fcd09cfb,
title = "Distributed prediction markets modeled by weighted Bayesian graphical games",
abstract = "We consider a novel, yet practical setting of prediction markets called distributed prediction markets, where the aggregated price of a security of an event in one prediction market is affected dynamically by the prices of securities of similar events in other, simultaneously running prediction markets. We focus on the problem of decision making facing a market maker to determine the price of a security within such a setting. We propose a formal framework based on graphical games called a weighted Bayesian graphical game (WBGG) to model the distributed prediction market setting and to capture the local interactions between multiple market makers. We then describe a distributed message passing algorithm based on NashProp algorithm to calculate the Bayes-Nash equilibrium in a WBGG. We provide analytical results including convergence and incentivizing truthful revelation among market makers. Our experimental results show that market makers that consider the influence of other market makers in a distributed prediction market setting while using our proposed WBGG-based algorithm obtain higher utilities and set prices more accurately in comparison to market makers using a greedy strategy to set prices or those that do not consider the influence of other market makers. We also observe that extreme sizes of the neighborhood of a market maker have an adverse impact on its utilities.",
keywords = "Bayes-Nash, Distributed, Graphical games, Prediction market",
author = "Janyl Jumadinova and Prithviraj Dasgupta",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/978-3-319-13218-1_7",
language = "English (US)",
volume = "187",
pages = "84--98",
journal = "Lecture Notes in Business Information Processing",
issn = "1865-1348",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Distributed prediction markets modeled by weighted Bayesian graphical games

AU - Jumadinova, Janyl

AU - Dasgupta, Prithviraj

PY - 2014/1/1

Y1 - 2014/1/1

N2 - We consider a novel, yet practical setting of prediction markets called distributed prediction markets, where the aggregated price of a security of an event in one prediction market is affected dynamically by the prices of securities of similar events in other, simultaneously running prediction markets. We focus on the problem of decision making facing a market maker to determine the price of a security within such a setting. We propose a formal framework based on graphical games called a weighted Bayesian graphical game (WBGG) to model the distributed prediction market setting and to capture the local interactions between multiple market makers. We then describe a distributed message passing algorithm based on NashProp algorithm to calculate the Bayes-Nash equilibrium in a WBGG. We provide analytical results including convergence and incentivizing truthful revelation among market makers. Our experimental results show that market makers that consider the influence of other market makers in a distributed prediction market setting while using our proposed WBGG-based algorithm obtain higher utilities and set prices more accurately in comparison to market makers using a greedy strategy to set prices or those that do not consider the influence of other market makers. We also observe that extreme sizes of the neighborhood of a market maker have an adverse impact on its utilities.

AB - We consider a novel, yet practical setting of prediction markets called distributed prediction markets, where the aggregated price of a security of an event in one prediction market is affected dynamically by the prices of securities of similar events in other, simultaneously running prediction markets. We focus on the problem of decision making facing a market maker to determine the price of a security within such a setting. We propose a formal framework based on graphical games called a weighted Bayesian graphical game (WBGG) to model the distributed prediction market setting and to capture the local interactions between multiple market makers. We then describe a distributed message passing algorithm based on NashProp algorithm to calculate the Bayes-Nash equilibrium in a WBGG. We provide analytical results including convergence and incentivizing truthful revelation among market makers. Our experimental results show that market makers that consider the influence of other market makers in a distributed prediction market setting while using our proposed WBGG-based algorithm obtain higher utilities and set prices more accurately in comparison to market makers using a greedy strategy to set prices or those that do not consider the influence of other market makers. We also observe that extreme sizes of the neighborhood of a market maker have an adverse impact on its utilities.

KW - Bayes-Nash

KW - Distributed

KW - Graphical games

KW - Prediction market

UR - http://www.scopus.com/inward/record.url?scp=84911957210&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84911957210&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-13218-1_7

DO - 10.1007/978-3-319-13218-1_7

M3 - Article

AN - SCOPUS:84911957210

VL - 187

SP - 84

EP - 98

JO - Lecture Notes in Business Information Processing

JF - Lecture Notes in Business Information Processing

SN - 1865-1348

ER -