Computing Equilibrium Wealth Distributions in Models with Heterogeneous-Agents, Incomplete Markets and Idiosyncratic Risk

This paper describes an accurate, fast and robust fixed point method for computing the stationary wealth distributions in macroeconomic models with a continuum of infinitely-lived households who face idiosyncratic shocks with aggregate certainty. The household wealth evolution is modeled as a mixtur...

Full description

Saved in:
Bibliographic Details
Published inComputational economics Vol. 41; no. 2; pp. 171 - 193
Main Authors Badshah, Muffasir, Beaumont, Paul, Srivastava, Anuj
Format Journal Article
LanguageEnglish
Published Boston Springer US 01.02.2013
Springer Nature B.V
Subjects
Algorithms
Behavioral/Experimental Economics
Capital stock
Computer Appl. in Social and Behavioral Sciences
Economic models
Economic Theory/Quantitative Economics/Mathematical Methods
Economics
Economics and Finance
Employment
Equilibrium
Expected utility
Households
Interest rates
Macroeconomics
Math Applications in Computer Science
Numerical analysis
Operations Research/Decision Theory
Studies
Unemployment
Wealth distribution
DSGE models
Stationary equilibria
Wealth distributions
Numerical solutions
Online AccessGet full text

Cover

More Information
Summary:This paper describes an accurate, fast and robust fixed point method for computing the stationary wealth distributions in macroeconomic models with a continuum of infinitely-lived households who face idiosyncratic shocks with aggregate certainty. The household wealth evolution is modeled as a mixture Markov process and the stationary wealth distributions are obtained using eigen structures of transition matrices by enforcing the conditions for the Perron–Frobenius theorem by adding a perturbation constant to the Markov transition matrix. This step is utilized repeatedly within a binary search algorithm to find the equilibrium state of the system. The algorithm suggests an efficient and reliable framework for studying dynamic stochastic general equilibrium models with heterogeneous agents.
ISSN:0927-7099
1572-9974
DOI:10.1007/s10614-011-9313-8