Abstract
Abstract Most of the many diagnostic tests for serial correlation (SC) and dynamic conditional heteroskedasticity (DCH) of regression disturbances have good asymptotic local power against finite autoregressive (AR) or moving average (MA) alternatives. In testing for SC, statistics typically depend on finitely many, p, least squares (LS) residual sample autocorrelations.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have