Abstract
A simple argument is used to derive the optimal GMM estimator of a finite order autoregressive process whose innovation may be conditionally heteroskedastic.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have