Abstract
In a formal statistical framework we examine the consequences of misspecifing sequential econometric models for their weak exogeneity properties. Modelling is viewed as a reduction sequence and the Bayesian paradigm is adopted. The central aim is inference on certain parameters of interest, which, in practice, almost invariably relies on some weak exogeneity assumption. Sufficient conditions for the preservation of weak exogeneity despite reduction errors are given, and several examples illustrate these findings. We conclude that some types of misspecification are much more dangerous than others, and we advocate attentive dynamic modelling.
Full Text
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call