Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/339345 
Year of Publication: 
2025
Series/Report no.: 
Working Paper No. 460
Version Description: 
Revised version, July 2025
Publisher: 
University of Zurich, Department of Economics, Zurich
Abstract: 
Economic agents often infer latent structures - such as preference types - from data, without exogenously specified priors. We model such agents as empirical Bayesians. They estimate both the prior over types and the meanings of types via maximum likelihood. We show this estimation is equivalent to decomposing the sample into subsamples, each best explained by a single available latent type, with the decomposition minimizing the average misfit. The equivalence yields structural properties: optimal latent representations are robust (type definitions locally invariant to data changes) and simple (type count bounded). We extend these properties to agents who face frictions in evaluating likelihoods.
Subjects: 
Bayesian updating
cognitive constraints
belief formation
machine learning in economics
Bayesian networks
Persistent Identifier of the first edition: 
Additional Information: 
This paper builds on and supersedes our previous paper, "Constrained Data-Fitters"
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.