NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Merging Simulation and Projection Approaches to Solve High-Dimensional Problems

Kenneth L. Judd, Lilia Maliar, Serguei Maliar

NBER Working Paper No. 18501
Issued in November 2012
NBER Program(s):   EFG   PE   TWP

We introduce an algorithm for solving dynamic economic models that merges stochastic simulation and projection approaches: we use simulation to approximate the ergodic measure of the solution, we construct a fixed grid covering the support of the constructed ergodic measure, and we use projection techniques to accurately solve the model on that grid. The grid construction is the key novel piece of our analysis: we select an ε-distinguishable subset of simulated points that covers the support of the ergodic measure roughly uniformly. The proposed algorithm is tractable in problems with high dimensionality (hundreds of state variables) on a desktop computer. As an illustration, we solve one- and multicountry neoclassical growth models and a large-scale new Keynesian model with a zero lower bound on nominal interest rates.

download in pdf format
   (975 K)

email paper

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w18501

Published: Merging simulation and projection approaches to solve high-dimensional problems with an application to a new Keynesian model Lilia Maliar1 andSerguei Maliar2,† Article first published online: 27 MAR 2015 DOI: 10.3982/QE364 Quantitative Economics Volume 6, Issue 1, pages 1–47, March 2015

Users who downloaded this paper also downloaded these:
Cai and Judd w18540 Dynamic Programming with Hermite Approximation
Judd, Maliar, and Maliar w15965 A Cluster-Grid Projection Method: Solving Problems with High Dimensionality
Judd, Maliar, Maliar, and Valero w19326 Smolyak Method for Solving Dynamic Economic Models: Lagrange Interpolation, Anisotropic Grid and Adaptive Domain
Cai, Judd, and Xu w18709 Numerical Solution of Dynamic Portfolio Optimization with Transaction Costs
Cai, Judd, Thain, and Wright w18714 Solving Dynamic Programming Problems on a Computational Grid
 
Publications
Activities
Meetings
NBER Videos
Themes
Data
People
About

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us