Identifying the Discount Factor in Dynamic Discrete Choice Models (with Øystein Daljord)
Abstract:
Empirical applications of stationary
dynamic discrete choice models usually either take the discount factor to
be known or rely on high level exclusion restrictions that are difficult to
interpret and hard to satisfy. We provide identification results under an
intuitively appealing exclusion restriction on primitive utilities that is
more directly useful to applied researchers. We show that such an exclusion
restriction, even with corresponding variation in future expected payoffs,
does not suffice for point identification. It does however identify the di
scount factor up to a discrete set, with only finitely many points outside
a neighborhood of one. Moreover, this identified set is easy to characteriz
e by solving a simple moment condition. We also show that our and existing
exclusion restrictions limit the choice and state transition probability da
ta in different ways; that is, they give the model nontrivial and distinct
empirical content.