Market Efficiency and Apparent Unit Roots: An Application to Exchange Rates

Prices in efficient markets are influenced by trading based on past patterns in the series. This induces parameter instability and near-random-walk behaviour in any time-series model of such data. Simulation results suggest that this parameter instability makes stationary series more likely to be erroneously classified as nonstationary, according to standard unit root or stationarity tests. It is shown that individual real exchange rate series appear individually non-stationary, especially for tests based on a null of stationarity, even though they appear stationary when treated as a panel.