Approximating and Forecasting Macroeconomic Signals in Real-Time

We optimally incorporate factors estimated from a large panel of macroeconomic time series in the estimation of two relevant signals related to real activity: business cycle fluctuations and the medium to long-run component of output growth. This latter signal conveys information on the growth of real activity but contains no high-frequency oscillations. For forecasting purposes we show that targeting this object can prove more useful than targeting the original (noisy) time series. We illustrate the methodology and provide forecasting comparisons for the Euro Area and the U.S.