Deep learning for solving dynamic economic models.

•We introduce a deep learning (DL) 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, Bellman equation and Euler equation.•We propose all-in-one integration technique that facilitates construction of high-dimensional expectation functions.•We use deep neural network to deal with multicollinearity and to perform model reduction.•Taken together, these techniques enable us to solve economic models with thousands of state variables, such as Krusell and Smith (1998) model.•We provide a TensorFlow code that accommodates a variety of applications.