Misspecified Beliefs
Paper Session
Friday, Jan. 3, 2025 8:00 AM - 10:00 AM (PST)
- Chair: Aislinn Bohren, University of Pennsylvania
Dynamic Concern for Misspecification
Abstract
We consider an agent who posits a set of probabilistic models for thepayoff-relevant outcomes. The agent has a prior over this set but fears the
actual model is omitted and hedges against this possibility. The concern for
misspecification is endogenous: If a model explains the previous
observations well, the concern attenuates. We show that different static
preferences under uncertainty (subjective expected utility, maxmin, robust
control) arise in the long run, depending on how quickly the agent becomes
unsatisfied with unexplained evidence and whether they are misspecified. The
misspecification concern's endogeneity naturally induces behavior cycles,
and we characterize the limit action frequency. This model is consistent
with the evidence on monetary policy cycles and choices in the face of
complex tax schedules. Finally, we axiomatize in terms of observable choices
this decision criterion and how quickly the agent adjusts their
misspecification concern.
Robust Comparative Statics with Misspecified Bayesian Learning
Abstract
We present novel monotone comparative statics results for steady state behavior in a dynamic optimization environment with misspecified Bayesian learning. We consider ageneralized framework, based on Esponda and Pouzo (2021), wherein a Bayesian learner facing a dynamic optimization problem has a prior on a set of parameterized transition
probability functions (models) but is misspecified in the sense that the true process is not within this set. In the steady state, the learner infers the model that best-fits the data generated by their actions, and in turn, their actions are optimally chosen given their inferred model. We characterize conditions on the primitives of the environment, and in particular, over the set of models under which the steady state distribution over
states and actions and inferred models exhibit monotonic behavior. Further, we offer a new theorem on the existence of a steady state on the basis of a monotonicity argument.
Lastly, we provide an upper bound on the cost of misspecification, again in terms of
the primitives of the environment. We demonstrate the utility of our results for several
environments of general interest, including forecasting models, dynamic effort-task, and
optimal consumption-savings problems
The Empirical Content of Bayesianism
Abstract
This paper characterizes the conditions under which the observed beliefs of a group of agents are consistent with Bayesian updating. Beliefs are consistent with Bayesianism if they arise from the application of Bayes’ rule given some subjective distribution for the state and the signals agents observe between periods. The paper’s main finding is that beliefs are consistent with Bayesianism if and only if the mean of the distribution of poste- riors is uniformly absolutely continuous with respect to the prior. Furthermore, the paper shows that the existing results on the empirical content of Bayesianism rely on additional restrictions on permissible subjective distributions, such as the requirement that agents have correct beliefs about the distribution of signals.JEL Classifications
- D8 - Information, Knowledge, and Uncertainty