Identification and estimation of dynamic structural models with unobserved choices

Abstract

This paper develops identification and estimation methods for dynamic discrete choice models when agents’ actions are unobserved by econometricians. We provide conditions under which choice probabilities and latent state transition rules are nonparametrically identified with a continuous state variable in a single-agent dynamic discrete choice model. Our identification strategy from the baseline model can extend to models with serially correlated unobserved heterogeneity, cases in which choices are partially unavailable, and dynamic discrete games. We propose a sieve maximum likelihood estimator for primitives in agents’ utility functions and state transition rules. Monte Carlo simulation results support the validity of the proposed approach.