Nested Pseudo Likelihood Estimation of Continuous-Time Dynamic Discrete Games

Jason R. Blevins and Minhae Kim.
The Ohio State University, Department of Economics
Working Paper.

Draft coming soon.

Abstract. This paper introduces a sequential estimation algorithm to estimate dynamic discrete choice models in continuous time. Dynamic discrete choice models have been used in various areas including dynamic demand models and firm entry-exit models which are mostly developed in discrete time setting. We extend the nested pseudo likelihood (NPL) estimator introduced by Aguirregabiria and Mira (2007) to continuous time models to better approximate the reality in certain economic contexts and to lessen the computational burden. We find that the NPL estimator in continuous time models has satisfying large sample properties as in discrete time models. Moreover, we present the local convergence condition in the iterative NPL algorithm and the zero Jacobian property assuring the local convergence in single agent models. Monte Carlo experiments using a five-player dynamic discrete game are executed to show the relative efficiency of the NPL estimator compared to two-step estimators.

Keywords: Continuous time, dynamic discrete games, dynamic discrete choice, nested pseudo likelihood.

BibTeX Record:

@TechReport{blevins-kim-2019,
author       = {Jason R. Blevins and Minhae Kim},
title        = {Nested Pseudo Likelihood Estimation of Continuous-Time
Dynamic Discrete Games},
institution  = {The Ohio State University},
year         = 2019,
type         = {Working Paper}
}