Abstract
Our results complement the recent findings of real exchange rates as stationary processes. Applying a battery of unit root tests can be problematic, since the tests are sensitive to the specifics of the time-series process. The novelty of our approach is in emphasizing the information content of the data to distinguish between the competing processes. Stationary and non-stationary ARIMA processes are fitted to the US/UK real exchange rate series, covering 134 years. Artificial data are generated, and the small sample distributions of the chosen test statistics are computed under each of the two hypotheses. The values of the actual sample statistics seem to come rather from the stationary than from the non-stationary process.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
