Limited-Information Estimation of Heterogeneous Agent Models

2026-08-14

AI Summary

This study addresses the limitation that local estimation in heterogeneous agent models typically relies on global assumptions by proposing a limited-information estimation method. By matching impulse responses and cross-sectional moments via sufficient statistics, this approach circumvents structural constraints on the remainder of the economy, enabling modular estimation and testing of model components. Applying this framework to a two-asset heterogeneous household model, we achieve parameter identification without imposing global restrictions. Consequently, this work effectively mitigates the sensitivity to misspecification inherent in traditional methods, providing a robust paradigm for the modular empirical analysis of complex macroeconomic models.

Abstract

We develop a method for estimating and testing a single block of a macroeconomic model with heterogeneous agents, without placing assumptions on the structure of the rest of the economy. In a large class of models, individual agents' decisions depend on the macroeconomy only through their expectations of the evolution of a finite-dimensional vector of "sufficient statistics" (e.g., asset returns or aggregate earnings). Our estimator selects the structural parameters that provide the best model-consistent fit between empirical impulse responses with respect to identified macro shocks of (a) cross-sectional moments of agent choices (e.g., moments of consumption) and (b) the vector of sufficient statistics. In a simulation illustration, we estimate a two-asset heterogeneous household model block without restricting production, firm investment, financial intermediation, monetary policy, trade, etc.

Problem

• Heterogeneous Agent Models
• Limited-Information Estimation
• Sufficient Statistics
• Macroeconomic Modeling

Innovation

• Limited-Information Estimation
• Heterogeneous Agent Models
• Sufficient Statistics
• Impulse Response Matching
• Cross-sectional Moments