Predicting human behavior in unrepeated, simultaneous-move games
From MaRDI portal
Abstract: It is common to assume that agents will adopt Nash equilibrium strategies; however, experimental studies have demonstrated that Nash equilibrium is often a poor description of human players' behavior in unrepeated normal-form games. In this paper, we analyze five widely studied models (Quantal Response Equilibrium, Level-k, Cognitive Hierarchy, QLk, and Noisy Introspection) that aim to describe actual, rather than idealized, human behavior in such games. We performed what we believe is the most comprehensive meta-analysis of these models, leveraging ten different data sets from the literature recording human play of two-player games. We began by evaluating the models' generalization or predictive performance, asking how well a model fits unseen test data after having had its parameters calibrated based on separate training data. Surprisingly, we found that what we dub the QLk model of Stahl & Wilson (1994) consistently achieved the best performance. Motivated by this finding, we describe methods for analyzing the posterior distributions over a model's parameters. We found that QLk's parameters were being set to values that were not consistent with their intended economic interpretations. We thus explored variations of QLk, ultimately identifying a new model family that has fewer parameters, gives rise to more parsimonious parameter values, and achieves better predictive performance.
Recommendations
- Level-0 models for predicting human behavior in games
- A Cognitive Hierarchy Model of Games
- A generalized cognitive hierarchy model of games
- Behavior in strategic settings: evidence from a million rock-paper-scissors games
- A dynamic graph model of strategy learning for predicting human behavior in repeated games
Cites work
- scientific article; zbMATH DE number 1907684 (Why is no real title available?)
- scientific article; zbMATH DE number 1912149 (Why is no real title available?)
- scientific article; zbMATH DE number 2117879 (Why is no real title available?)
- scientific article; zbMATH DE number 3106184 (Why is no real title available?)
- A Cognitive Hierarchy Model of Games
- A Simplex Method for Function Minimization
- A cognitive hierarchy model of learning in networks
- A dynamic homotopy interpretation of the logistic quantal response equilibrium correspondence
- A model of noisy introspection.
- Action-graph games
- An experimental investigation of unprofitable games.
- Equilibrium play and best response to (stated) beliefs in normal form games
- Evidence for optimistic and pessimistic behavior in normal-form games
- Evidence on the equivalence of the strategic and extensive form representation of games.
- Experience-weighted Attraction Learning in Normal Form Games
- Heterogeneous quantal response equilibrium and cognitive hierarchies
- How portable is level-0 behavior? A test of level-\(k\) theory in games with non-neutral frames
- Ignoring the rationality of others: evidence from experimental normal-form games.
- Improvements on Cross-Validation: The .632+ Bootstrap Method
- Level-k Auctions: Can a Nonequilibrium Model of Strategic Thinking Explain the Winner's Curse and Overbidding in Private-Value Auctions?
- Machine learning. A probabilistic perspective
- Multiagent Systems
- On players' models of other players: Theory and experimental evidence
- On the beliefs off the path: equilibrium refinement due to quantal response and level-\(k\)
- On the persistence of strategic sophistication
- Out of your mind: eliciting individual reasoning in one shot games
- Pattern recognition and machine learning.
- Quantal response and nonequilibrium beliefs explain overbidding in maximum-value auctions
- Quantal response equilibria for normal form games
- Stated Beliefs and Play in Normal-Form Games
- Strategic reasoning in p-beauty contests
- Strategic sophistication and attention in games: an eye-tracking study
Cited in
(11)- Games with second-order expected utility
- A computational model of outguessing in two-player non-cooperative games
- scientific article; zbMATH DE number 7558435 (Why is no real title available?)
- The Hanabi challenge: a new frontier for AI research
- Behavior in strategic settings: evidence from a million rock-paper-scissors games
- A case of evolutionarily stable attainable equilibrium in the laboratory
- Level-0 models for predicting human behavior in games
- scientific article; zbMATH DE number 18237 (Why is no real title available?)
- A dynamic graph model of strategy learning for predicting human behavior in repeated games
- A comparison of naive and sophisticated subject behavior with game theoretic predictions
- Comparing theories of one-shot play out of treatment
This page was built for publication: Predicting human behavior in unrepeated, simultaneous-move games
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1682705)
