Games with Incomplete Information Played by ‘Bayesian’ Players, Part III. The Basic Probability Distribution of the Game

From MaRDI portal
Revision as of 03:36, 7 March 2024 by Import240305080351 (talk | contribs) (Created automatically from import240305080351)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Publication:5564028

DOI10.1287/MNSC.14.7.486zbMath0177.48501OpenAlexW2044155094MaRDI QIDQ5564028

John C. Harsanyi

Publication date: 1968

Published in: Management Science (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1287/mnsc.14.7.486




Related Items (only showing first 100 items - show all)

AGENT CONNECTEDNESS AND BACKWARD INDUCTIONTHE PEDESTRIAN PRINCIPLE FOR DIFFERENTIAL GAMESAlmost common priorsNetwork routing for insurgency: An adversarial risk analysis frameworkPerfect Bayesian equilibrium and sequential equilibriumContinuous implementation of constrained rational expectations equilibriaRational behavior with payoff uncertaintyUncertain information structures and backward inductionGames with Symmetric Incomplete Information and Asymmetric Computational ResourcesA Rough View on Incomplete Information in GamesUnnamed ItemData envelopment analysis approaches for two-level production and distribution planning problemsNonlocal correlations and noise in different settings of a two-player gamePerfect regular equilibriumModeling, analysis, and dynamics of Bayesian games via matrix-based methodAn epistemic approach to stochastic gamesThe existence of universal qualitative belief spacesSome Remarks on Bayesian Mechanism DesignConjectures and underpricing in repeated mass disputes with heterogeneous plaintiffsCoordinated replenishment game and learning under time dependency and uncertainty of the parametersContinuous level-\(k\) mechanism designAgreeing to DisagreeEpistemic Conditions for Nash EquilibriumComputational Efficiency in Multivariate Adversarial Risk Analysis ModelsOn the existence of pure-strategy equilibria in games with private information: a complete characterizationUnnamed ItemOne-sided adaptation for infinite-horizon linear quadraticN-person non-zero-sum dynamic games and sensitivity analysisStatic GamesBayesian Games: Games with Incomplete InformationRepeated Games with Incomplete InformationFROM CLASSICAL TO EPISTEMIC GAME THEORYUNIVERSALLY RATIONAL BELIEF HIERARCHIESON THE COMPLETENESS OF THE UNIVERSAL KNOWLEDGE–BELIEF SPACE: A TECHNICAL NOTECountable spaces and common priorsUniversal knowledge-belief structuresThe information operation in a repeated n-person nonzero-sum gameThe coming of game theoryThe non-existence of a universal topological type spaceBelief-independence and (robust) strategy-proofnessPeriodic strategies. II: Generalizations and extensionsList of Publications: John C. Harsanyi, December 1997Conditional Probability Logic over Conditional Probability SpacesThe return function: a new computable perspective on Bayesian-Nash equilibriaInequality and Procedural Justice in Social DilemmasEnvironmental game modeling with uncertaintiesUnnamed ItemRisk management in hierarchical games with random factorsAdversarial risk analysis: Borel gamesAmbiguous language and common priorsA parametric family of two ranked objects auctions: equilibria and associated riskThe value of information in a simple duopoly modelSome extensions of a claim of Aumann in an axiomatic model of knowledgeImplementation in differential information economiesBelief and truth in hypothesised behavioursElections with limited information: A fulfilled expectations model using contemporaneous poll and endorsement data as information sourcesOptimal research for Cournot oligopolistsInfrastructure security gamesDistributed algorithms for the computation of noncooperative equilibriaWho wants to break the hockey-stick sales pattern in the supply chain?A new theory of equilibrium selection for games with complete informationAn axiomatic theory of a risk dominance measure for bipolar games with linear incentivesGames with incomplete information when players are partially aware of others' signalsDynamic consistency in incomplete information games under ambiguityAutonomous agents modelling other agents: a comprehensive survey and open problemsBayesian learning and convergence to rational expectationsCommon knowledge with probability 1Bayesian Nash equilibria using extended Werner-like statesReputation in repeated second-price auctionsPsychology of game playing: introduction to a special issueThe Bayesian foundations of solution concepts of gamesVirtual implementation in incomplete information environments with infinite alternatives and typesThe work of John F. Nash Jr. in game theory. Nobel seminar, 8 December 1994An early paper on the refinement of Nash equilibriumEquilibria of the sealed-bid mechanism for bargaining with incomplete informationStatic game theory models and their applications in management scienceBilateral trade with the sealed bid k-double auction: Existence and efficiencyThe rate at which a simple market converges to efficiency as the number of traders increases: An asymptotic result for optimal trading mechanismsA robust optimization approach for solving two-person games under interval uncertaintyNear-Nash equilibrium strategies for LQ differential games with inaccurate state informationOptimal allocation of MRI scan capacity among competing hospital departmentsAsymmetric auction experiments with(out) commonly known beliefsAgency equilibriumRobust games: theory and application to a Cournot duopoly modelStackelberg-solvable games and pre-play communicationKnightian games and robustness to ambiguityIrregular gaming areasThree-player conflicting interest games and nonlocalityCoalitional game with fuzzy payoffs and credibilistic coreThe \textit{ex ante} \(\alpha\)-core for normal form games with uncertaintyA model of electoral competition with incomplete informationCoherent behavior in noncooperative gamesSimultaneous offers and the inefficiency of bargaining: A two-period exampleOptimal coordination mechanisms in generalized principal-agent problemsBayesian incentive compatible beliefsSemidefinite complementarity reformulation for robust Nash equilibrium problems with Euclidean uncertainty setsPredation, reputation, and entry deterrenceReputation and imperfect informationCoherent beliefs are not always types.Modelling incomplete information in Boolean games using possibilistic logicDynamic pricing of call rates: Bayesian approach







This page was built for publication: Games with Incomplete Information Played by ‘Bayesian’ Players, Part III. The Basic Probability Distribution of the Game