The Beveridge–Nelson decomposition calculates trend and cycle for an integrated time series. However, there are two ways to interpret the results from the decomposition. One interpretation is that the optimal long-run forecast (minus any deterministic drift) used to calculate the Beveridge–Nelson trend corresponds to an estimate of an unobserved permanent component. The other interpretation is that the optimal long-run forecast defines an observable permanent component. This paper examines some issues surrounding these two interpretations and provides empirical support for interpreting the Beveridge–Nelson trend as an estimate when considering macroeconomic data.
