| Abstract: |
Engel and West (EW, 2005) argue that as the discount factor gets closer to
one, present-value asset pricing models place greater weight on future
fundamentals. Consequently, current fundamentals have very weak forecasting
power and exchange rates appear to follow approximately a random walk. We
connect the Engel-West explanation to the studies of exchange rates with
long-horizon regressions. We find that under EW's assumption that fundamentals
are I(1) and observable to the econometrician, long-horizon regressions
generally do not have significant forecasting power. However, when EW's
assumptions are violated in a particular way, our analytical results show that
there can be substantial power improvements for long-horizon regressions, even
if the power of the corresponding shorthorizon regression is low. We simulate
population Rsquared for long-horizon regressions in the latter setting, using
Monetary and Taylor Rule models of exchange rates calibrated to the data.
Simulations show that long-horizon regression can have substantial forecasting
power for exchange rates. |