Testing for Unit Roots in Exchange Rates
本文检验了后布雷顿森林体系时期实际汇率是否具有单位根，发现拒绝单位根假设的证据主要指向爆炸性替代假设，而非稳定替代假设，表明低检验功效可能不是无法拒绝单位根假设的主要原因。
It is sometimes suggested that the failure of statistical tests to reject the null hypothesis of a unit root in post Bretton- Woods real exchange rates is a consequence of the low power of these tests. In this paper we address this methodological issue by examining both stationary and explosive alternatives to a unit-root null using procedures developed in Bhargava (1986). We show that the evidence which suggests rejection of a unit root does so predominantly in the direction of an explosive alternative-in the opposite direction of a stable alternative. This suggests that low test power may not explain well the failure to reject a unit-root null.