Linear IV regression estimators for structural dynamic discrete choice models

In structural dynamic discrete choice models, unobserved or mis-measured state variables may lead to biased parameter estimates and misleading inference. this paper, we show that instrumental can address such measurement problems when they relate evolve exogenously from the perspective of individual agents (i.e., market-level states). We define a class linear estimators rely on Euler equations expressed in terms conditional probabilities (ECCP estimators). These do not require observing modeling agent’s entire information set, nor solving simulating program. As such, are simple implement computationally light. provide constructive arguments for identification model primitives, establish estimator’s consistency asymptotic normality. Four applied examples serve illustrate ECCP approach’s implementation, advantages, limitations: demand durable goods, agricultural land use change, technology adoption, labor supply. good finite-sample performance Monte Carlo study, estimate supply empirically taxi drivers New York City.