Forecasting Inflation With a Random Walk

The use of different time-series models to generate forecasts is fairly usual in the forecasting literature in general, and in the inflation forecast literature in particular. When the predicted variable is stationary, the use of processes with unit roots may seem counterintuitive. Nevertheless, in this paper we demonstrate that forecasting a stationary variable with driftless unit-root-based forecasts generates bounded Mean Squared Prediction Errors errors at every single horizon. We also show via simulations that persistent stationary processes may be better predicted by unit-root-based forecasts than by forecasts coming from a model that is correctly specified but that is subject to a higher degree of parameter uncertainty. Finally we provide an empirical illustration in the context of CPI inflation forecasts for three industrialized countries.