This paper defines a class of linear instrumental variables estimators for structural dynamic discrete choice models that rely on Euler equations expressed in conditional choice probabilities (hence, ECCP estimators).  These estimators are simple to implement, are compatible with serially correlated market-level unobservables and do not require fully observed state models.  Like static reduced-form estimators, ECCP estimators admit the use of instrumental variables to deal with endogeneity problems.  We provide constructive identification arguments and establish consistency and asymptotic normality.  A Monte Carlo study demonstrates the finite sample performance of ECCP estimators in the context of a model of dynamic demand for durable goods.

Presented by Paul Scott, New York University

About Econometrics Colloquium Seminar Series

A seminar series designed specifically for econometricians to network and collaborate.


Colin Clark Building (#39)