Abstract
The basic theorem characterizing Granger causality events and existing testing methods are surveyed. An alternative direct testing method based on Akaike's final prediction error criterion is suggested. The various methods are compared both empirically and theoretically by applying them to postwar U.S. money and income data and by discussing the power of the various tests for some simple models.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have