Computer Code
Link to Victor Aguirregabiria's Library of Gauss Procedures for Econometrics and Computational Methods in Economics, with examples of programs that call the procedures.
Link to Christopher Ferrall's GitHub repository of papers with empirical applications of Dynamic Discrete Choice Structural models. It includes code and data.
Link to Eric Schulman's Python code for the implementation of the Nested Pseudo Likelihood method in Aguirregabiria and Mira (ECMA, 2020).
Link to Patrick Mogesen's Julia code for the solution of Rust's bus engine replacement model using Gaussian Quadrature
Link to Jaap Abbring and Tobias Klein's Matlab code on Econometric Methods for Dynamic Discrete Choice models.
COMPANION WEB PAGE OF THE PAPER
“DYNAMIC DISCRETE CHOICE STRUCTURAL MODELS: A SURVEY”
BY VICTOR AGUIRREGABIRIA AND PEDRO MIRA
COMPUTER CODE
This web page contains computer code which implements methods for the estimation of discrete choice dynamic programming models (single agent models, competitive equilibrium models and dynamic games). These methods are reviewed (with more or less detail) in the survey paper “Dynamic Discrete Choice Structural Models: A Survey,” by Victor Aguirregabiria and Pedro Mira.
The computer programs included below have been generously provided by the authors of the methodological papers that first proposed each method. The authors have contributed ZIP files with programs and documentation (readme files), and in some case with datasets. We have just posted these ZIP files on this webpage.
Aguirregabiria and Mira: Nested Pseudo Likelihood (NPL)
Arcidiacono and Jones / Arcidiacono and Miller: Sequential EM Algorithm
Bajari, Benkard and Levin: BBL
Imai, Jain and Ching: Bayesian Estimation
Keane and Wolpin: Simulation and Interpolation
Lee and Wolpin: Dynamic Competitive Equilibrium
Pakes, Ostrovsky and Berry: Dynamic Oligopoly Entry-Exit Game
Rust: Nested Fixed Point Algorithm (NFXP)