Limit Points of Endogenous Misspecified Learning

We study how an agent learns from endogenous data when their prior belief is misspecified. We show that only uniform Berk–Nash equilibria can be long‐run outcomes, and that all uniformly strict Berk–Nash equilibria have an arbitrarily high probability of being the long‐run outcome for some initial beliefs. When the agent believes the outcome distribution is exogenous, every uniformly strict Berk–Nash equilibrium has positive probability of being the long‐run outcome for any initial belief. We generalize these results to settings where the agent observes a signal before acting.

[1]  Alwyn Young,et al.  Misspecified Politics and the Recurrence of Populism , 2022, American Economic Review.

[2]  D. Fudenberg,et al.  Pathwise Concentration Bounds for Misspecified Bayesian Beliefs , 2021, SSRN Electronic Journal.

[3]  Kevin He,et al.  Evolutionarily Stable (Mis)specifications: Theory and Applications , 2020, EC.

[4]  Ignacio Esponda,et al.  Asymptotic behavior of Bayesian learners with misspecified models , 2019, J. Econ. Theory.

[5]  Ignacio Esponda,et al.  Equilibrium in Misspecified Markov Decision Processes , 2015, Theoretical Economics.

[6]  L. Hartmann.  α-Maxmin Expected Utility with Non-Unique Prior , 2021 .

[7]  B. Kőszegi,et al.  Convergence in models of misspecified learning , 2021 .

[8]  D. Fudenberg,et al.  Which Misperceptions Persist? , 2020 .

[9]  Ryota Iijima,et al.  Misinterpreting Others and the Fragility of Social Learning , 2019, Econometrica.

[10]  R. Spiegler,et al.  A Model of Competing Narratives , 2018, American Economic Review.

[11]  Drew Fudenberg,et al.  Payoff information and learning in signaling games , 2020, Games Econ. Behav..

[12]  G. Mailath,et al.  Learning under Diverse World Views: Model-Based Inference , 2019, American Economic Review.

[13]  Paul Heidhues,et al.  Overconfidence and Prejudice , 2019, 1909.08497.

[14]  Dmitry Taubinsky,et al.  Rules of Thumb and Attention Elasticities: Evidence from Under- and Overreaction to Taxes , 2019, The Review of Economics and Statistics.

[15]  J. V. Reenen,et al.  Joseph Schumpeter Lecture, EEA Annual Congress 2017: Do Tax Cuts Produce more Einsteins? The Impacts of Financial Incentives VerSus Exposure to Innovation on the Supply of Inventors , 2019, Journal of the European Economic Association.

[16]  Raj Chetty,et al.  Do Tax Cuts Produce More Einsteins? The Impacts of Financial Incentives vs. Exposure to Innovation on the Supply of Inventors , 2019 .

[17]  Dmitry Taubinsky,et al.  Measuring “Schmeduling” , 2016, The Review of Economic Studies.

[18]  Pooya Molavi.  Macroeconomics with Learning and Misspecification: A General Theory and Applications , 2019 .

[19]  Philipp Strack,et al.  Convergence in Misspecified Learning Models with Endogenous Actions , 2018 .

[20]  P. Jehiel.  Investment Strategy and Selection Bias: An Equilibrium Perspective on Overoptimism , 2018, American Economic Review.

[21]  M. Rabin,et al.  Channeled Attention and Stable Errors , 2018 .

[22]  Kevin He,et al.  Mislearning from Censored Data: The Gambler's Fallacy in Optimal-Stopping Problems , 2018 .

[23]  D. Fudenberg,et al.  Player-Compatible Equilibrium , 2017, 1712.08954.

[24]  In-Koo Cho,et al.  Gresham's Law of Model Averaging , 2017 .

[25]  Drew Fudenberg,et al.  Active learning with a misspecified prior , 2017 .

[26]  Massimo Warglien,et al.  What Situation Is This? Coarse Cognition and Behavior Over a Space of Games , 2017 .

[27]  J. Aislinn Bohren,et al.  Bounded Rationality and Learning: A Framework and a Robustness Result , 2017 .

[28]  Philipp Strack,et al.  Unrealistic Expectations and Misguided Learning , 2017 .

[29]  D. Fudenberg,et al.  Learning and Type Compatibility in Signaling Games , 2017, 1702.01819.

[30]  D. Fudenberg,et al.  Bayesian posteriors for arbitrarily rare events , 2016, Proceedings of the National Academy of Sciences.

[31]  J. Aislinn Bohren,et al.  Informational Herding with Model Misspecification , 2013, J. Econ. Theory.

[32]  Tristan Gagnon-Bartsch.  Taste Projection in Models of Social Learning ∗ , 2016 .

[33]  Ignacio Esponda,et al.  An Equilibrium Framework for Players with Misspecified Models , 2014, ArXiv.

[34]  Ignacio Esponda,et al.  Berk-Nash Equilibrium: A Framework for Modeling Agents with Misspecified Models , 2014, 1411.1152.

[35]  Ignacio Esponda,et al.  Information feedback in first price auctions , 2008 .

[36]  Philippe Jehiel,et al.  Revisiting games of incomplete information with analogy-based expectations , 2008, Games Econ. Behav..

[37]  A. Rustichini,et al.  Representing preferences with a unique subjective state space: A corrigendum , 2007 .

[38]  L. Kogan,et al.  The Gambler’s and Hot-Hand Fallacies: Theory and Applications , 2007 .

[39]  Kenneth Kasa,et al.  Learning and Model Validation , 2007 .

[40]  Philippe Jehiel,et al.  Analogy-based expectation equilibrium , 2004, J. Econ. Theory.

[41]  Z. Yang,et al.  Probability models for DNA sequence evolution , 2004, Heredity.

[42]  Barton L. Lipman,et al.  REPRESENTING PREFERENCES WITH A UNIQUE SUBJECTIVE STATE SPACE , 2001 .

[43]  Eld,et al.  Cursed Equilibrium , 2000 .

[44]  M. Hirsch,et al.  Mixed Equilibria and Dynamical Systems Arising from Fictitious Play in Perturbed Games , 1999 .

[45]  J. Kagel,et al.  The Winner's Curse and Public Information in Common Value Auctions: Comment , 1999 .

[46]  Joel L. Schrag,et al.  First Impressions Matter: A Model of Confirmatory Bias , 1999 .

[47]  Nicholas C. Yannelis,et al.  Bounded rationality and learning , 1994 .

[48]  Robert J. Plemmons,et al.  Nonnegative Matrices in the Mathematical Sciences , 1979, Classics in Applied Mathematics.

[49]  Kim C. Border,et al.  Infinite Dimensional Analysis: A Hitchhiker’s Guide , 1994 .

[50]  David M. Kreps,et al.  Learning Mixed Equilibria , 1993 .

[51]  D. Fudenberg,et al.  Maintaining a Reputation when Strategies are Imperfectly Observed , 1992 .

[52]  Yaw Nyarko,et al.  Learning In Mis-Specified Models And The Possibility Of Cycles , 1991 .

[53]  D. Freedman,et al.  On the uniform consistency of Bayes estimates for multinomial probabilities , 1990 .

[54]  Dudley,et al.  Real Analysis and Probability: Measurability: Borel Isomorphism and Analytic Sets , 2002 .

[55]  I. Gilboa,et al.  Maxmin Expected Utility with Non-Unique Prior , 1989 .

[56]  M. Bray,et al.  Rational Expectations Equilibria, Learning, and Model Specification , 1986 .

[57]  M. Bray.  Learning, estimation, and the stability of rational expectations , 1982 .

[58]  A. Tversky,et al.  Evidential impact of base rates , 1981 .

[59]  J. Neveu,et al.  Discrete Parameter Martingales , 1975 .

[60]  Daniel Kahneman,et al.  Availability: A heuristic for judging frequency and probability , 1973 .

[61]  Kenneth J. Arrow,et al.  Notes on Expectations Equilibria in Bayesian Settings Institute for Mathematical Studies in the Social Sciences , 1973 .

[62]  R. Berk,et al.  Limiting Behavior of Posterior Distributions when the Model is Incorrect , 1966 .

[63]  Simon Kochen,et al.  A note on the Borel-Cantelli lemma , 1964 .

[64]  D. Ellsberg.  Decision, probability, and utility: Risk, ambiguity, and the Savage axioms , 1961 .

[65]  O. Gaans.  Probability measures on metric spaces , 2022 .