Dynamic Discrete Choice Models with Proxies for Unobserved Technologies

Abstract

Firms make forward-looking decisions based on latent technological states. While the true state is not observed by econometricians, the literature provides ways to construct proxies. For dynamic discrete choice models of forward-looking firms where a continuous state variable is unobserved but its proxy is available, we derive closed-form identification of the conditional choice probability, the Markov law of state transition, and the underlying structural parameters by explicitly solving relevant integral equations. We use this method to estimate the structures of firms and the option values of exit across industries.