We argue that deep learning provides a promising avenue for taming the curse of dimensionality in quantitative economics. We begin by exploring the unique challenges posed by solving dynamic equilibrium models, especially the feedback loop between individual agents' decisions and the aggregate consistency conditions required by equilibrium. Following this, we introduce deep neural networks and demonstrate their application by solving the stochastic neoclassical growth model. Next, we compare deep neural networks with traditional solution methods in quantitative economics. We conclude with a survey of neural network applications in quantitative economics and offer reasons for cautious optimism.
Authors
Related Organizations
- Acknowledgements & Disclosure
- This paper is based on Jesús Fernández-Villaverde’s keynote address at the 2024 Econometric Society Interdisciplinary Frontiers (ESIF) conference on Economics and AI+ML at Cornell. We are very grateful to Douglas de Araujo, Matthias Rottner, Simon Scheidegger, and Yucheng Yang for their comments. The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the Banco de España or the Eurosystem. All remaining errors are ours. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
- DOI
- https://doi.org/10.3386/w33117
- Pages
- 50
- Published in
- United States of America
Table of Contents
- Introduction 3
- Solving DEMs 7
- Neural networks 9
- Neural networks as function approximators 9
- Training the NN 13
- Designing the NN architecture 17
- Why do NNs work? 18
- An example: The stochastic growth model 20
- Applications of deep NNs in DEMs 27
- Dynamic programming 27
- Heterogeneous-agent models 29
- Reinforcement learning 32
- Estimation 33
- Traditional methods vs. NNs 34
- Reasons for cautious optimism 37
- Reason 1: Low-dimensional representations 37
- Reason 2: The manifold hypothesis 38
- Reason 3: Transferability between models and invariants 39
- A glimpse into the future 40
