Project
Set identification in nonlinear panel data models.
Panel data models are used in a wide range of economic applications, but in models with unobserved heterogeneity, estimates of the structural parameters will in general be inconsistent. This is referred to as the incidental parameters problem. Recently, it has been shown that in some models point identification may not even be possible. Given the failure of point identification, it is natural to shift the attention from point estimation to calculation and estimation of the identified set. The aim of this Ph.D. project is twofold. Firstly, to develop and improve methods to construct the identified set. Secondly, to investigate the relation between approximate solutions for the incidental parameter problem and the identified set. The focus will be on dynamic binary panel data models.