Entropy Regularized Belief Reporting

TL;DR

This paper introduces the Entropy Regularized Belief Reporting model, explaining how agents report beliefs based on a latent prior and entropy considerations, with applications to experimental data.

Contribution

The paper develops a novel model capturing partition dependence in belief reporting using entropy regularization and provides identification methods for the latent prior.

Findings

Model successfully explains partition dependence in belief reports.

Application to experimental data demonstrates the model's empirical relevance.

Provides structural properties for identifying latent priors.

Abstract

This paper investigates a model of partition dependence, a widely reported experimental finding where the agent's reported beliefs depend on how the states are grouped. In the model, called Entropy Regularized Belief Reporting (ERBR), the agent is endowed with a latent benchmark prior that is unobserved by the analyst. When presented with a partition, the agent reports a prior that minimizes Kullback-Leibler divergence from the latent benchmark prior subject to entropy regularization. This captures the intuition that while the agent would like to report a prior that is close to her latent benchmark prior, she may also have a preference to remain noncommittal. I provide the structural properties of the model that allow for identification of the latent benchmark prior and apply the model to the experimental data from Benjamin et al. (2017).