Mislearning from censored data: The gambler's fallacy and other correlational mistakes in optimal‐stopping problems

@article{He2022MislearningFC,
  title={Mislearning from censored data: The gambler's fallacy and other correlational mistakes in optimal‐stopping problems},
  author={Kevin He},
  journal={Theoretical Economics},
  year={2022}
}
  • Kevin He
  • Published 21 March 2018
  • Economics
  • Theoretical Economics
I study endogenous learning dynamics for people who misperceive intertemporal correlations in random sequences. Biased agents face an optimal‐stopping problem. They are uncertain about the underlying distribution and learn its parameters from predecessors. Agents stop when early draws are “good enough,” so predecessors' experiences contain negative streaks but not positive streaks. When agents wrongly expect systematic reversals (the “gambler's fallacy”), they understate the likelihood of… 
Subjective Causality in Choice
An agent makes a stochastic choice from a set of lotteries. She infers the outcomes of each available lottery using a subjective causal model represented by a directed acyclic graph, and consequently

References

SHOWING 1-10 OF 45 REFERENCES
The Gambler's and Hot-Hand Fallacies: Theory and Applications
We develop a model of the gambler's fallacy (the mistaken belief that random sequences should exhibit systematic reversals). We show that an individual who holds this belief and observes a sequence
Learning with misattribution of reference dependence
Informational Herding with Model Misspecification
TLDR
It is shown that when individuals significantly overestimate the amount of new information contained in prior actions, beliefs about the unknown state become entrenched and incorrect learning may occur, and when agents have an approximately correct model of inference, learning is complete.
Which Misperceptions Persist?
We use an evolutionary model to determine what sorts of misperceptions are likely to persist. In our model, every period a new generation of agents uses their subjective models and the outcome
Convergence in models of misspecified learning
We establish convergence of beliefs and actions in a class of one-dimensional learning settings in which the agent’s model is misspecified, she chooses actions endogenously, and the actions affect
Naïve Herding in Rich-Information Settings
In social-learning environments, we investigate implications of the assumption that people naively believe that each previous person's action reflects solely that person's private information. Naive
Belief Convergence under Misspecified Learning: A Martingale Approach
TLDR
This work introduces a novel “prediction accuracy” order over subjective models, and derives general conditions to determine when beliefs in a given environment converge to some long-run belief either locally or globally.
Biased Beliefs About Random Samples: Evidence from Two Integrated Experiments
This paper describes results of a pair of incentivized experiments on biases in judgments about random samples. Consistent with the Law of Small Numbers (LSN), participants exaggerated the likelihood
Everyday econometricians: Selection neglect and overoptimism when learning from others
In this paper, we design an investment game which allows us to study the influence of selection when learning from others. Using the theoretical study of selection neglect in Jehiel (2018) as a
...
...