RECURSIVE INATTENTIVENESS WITH HETEROGENEOUS EXPECTATIONS

Abstract

We model agents' endogenous updating of information sets over time under changing macroeconomic conditions. Building on sticky information models, the degree of inattentiveness is endogenized by allowing agents to choose between a costly full-information predictor and a costless sticky-information predictor. This is modeled as a choice between discrete alternatives under rational inattention. Recursive simulation shows that the dynamic equilibrium paths of aggregate variables are highly persistent and match the moments of U.S. data better than a model with fixed sticky information or with sticky prices, especially with regard to higher moments and the degree of persistence. Predictors are chosen in line with the predictions from rational inattention models, as the aggregate degree of attentiveness increases with rising variance of the forecast variable. Moreover, the model can generate hump-shaped impulse responses of inflation to a monetary policy shock if the degree of inattentiveness is sufficiently high.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.