Using the Yield Curve in Forecasting Output Growth and In?flation

Following Diebold and Li (2006), we use the Nelson-Siegel (NS, 1987) yield curve factors. However the NS yield curve factors are not supervised for a specifi?c forecast target in the sense that the same factors are used for forecasting different variables, e.g., output growth or infl?ation. We propose a modifed NS factor model, where the new NS yield curve factors are supervised for a specifi?c variable to forecast. We show it outperforms the conventional (non-supervised) NS factor model in out-of-sample forecasting of monthly US output growth and infl?ation. The original NS yield factor model is to combine information (CI) of predictors and uses factors of predictors (yield curve). The new supervised NS factor model is to combine forecasts (CF) and uses factors of forecasts of output growth or infl?ation conditional on the yield curve. We formalize the concept of supervision, and demonstrate analytically and numerically how supervision works. For both CF and CI schemes, principal components (PC) may be used in place of the NS factors. In out-of-sample forecasting of U.S. monthly output growth and infl?ation, we fi?nd that supervised CF-factor models (CF-NS, CF-PC) are substantially better than unsupervised CI-factor models (CI-NS, CI-PC), especially at longer forecast horizons.