ABSTRACT
Curiously enough, such a well-known equilibrium theory as uncovered interest parity is quite often misinterpreted in econometric analysis. One should remember that it is a one-period model; and great care is needed when specifying precisely what happens at the end of that period and at its beginning. One-period models in economics are actually very tricky. Or, to put the same point in another way: uncovered interest parity is a difference equation, in continuous time a differential equation. And in solving differential equations one has to be careful to specify initial conditions, and not just forget them. Surprisingly, even such an obvious and elementary point needs to be emphasized.
