This paper investigates whether the Markov switching model is a useful
tool for describing the behavior of floating exchange rates more generally.
The time-series properties of 18 exchange rates in the post-Bretton Woods
period, including 11 non-U.S. dollar exchange rates, are investigated. In
general, the model of Engel and Hamilton (EH) does not clearly outperform
the random walk model or the forward exchange rate in out-of-sample
forecasts. The mean squared errors of the forecasts of the segmented trends
model tend not to be significantly lower than those of a zero-drift random
walk, a random walk with drift or the forward rate. There is some evidence,