We argue that deep learning offers a promising way to manage the curse of scale in economies of scale. We begin by examining the unique challenges posed by solving dynamic equilibrium models, particularly the feedback loop between individual agent decisions and the overall consistency conditions required for equilibrium. Following this, we introduce deep neural networks and demonstrate their use by solving a 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 provide reasons for cautious optimism.
That’s according to a new paper by Jesús Fernández-Villaverde, Galo Nuño, and Jesse Perla.
Source link