This paper reconsiders several recently published but controversial results about the behaviour of exchange rates. In particular, it explores finite-sample problems in the application of cointegration tests and shows how these may have affected the conclusions of recent research.
Several recent papers have presented evidence from foreign exchange and other markets suggesting that the log of excess returns can be characterized as first-order integrated processes (I(1)). This contrasts sharply with the "conventional" wisdom that log prices are integrated of order one I(1) and that log returns should therefore be integrated of order zero I(0), and even more sharply with the view that past returns have no ability to predict future returns (weak market efficiency).
