Deep learning for solving dynamic economic models.

A-Tier
Journal: Journal of Monetary Economics
Year: 2021
Volume: 122
Issue: C
Pages: 76-101

Authors (3)

Maliar, Lilia (not in RePEc) Maliar, Serguei (Santa Clara University) Winant, Pablo (not in RePEc)

Score contribution per author:

1.341 = (α=2.01 / 3 authors) × 2.0x A-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

We introduce a unified deep learning method that solves dynamic economic models by casting them into nonlinear regression equations. We derive such equations for three fundamental objects of economic dynamics – lifetime reward functions, Bellman equations and Euler equations. We estimate the decision functions on simulated data using a stochastic gradient descent method. We introduce an all-in-one integration operator that facilitates approximation of high-dimensional integrals. We use neural networks to perform model reduction and to handle multicollinearity. Our deep learning method is tractable in large-scale problems, e.g., Krusell and Smith (1998). We provide a TensorFlow code that accommodates a variety of applications.

Technical Details

RePEc Handle
repec:eee:moneco:v:122:y:2021:i:c:p:76-101
Journal Field
Macro
Author Count
3
Added to Database
2026-01-26