Title: Boosting non-linear predictability of macroeconomic time series
Authors: Timo Virtanen - University of Turku (Finland) [presenting]
Heikki Kauppi - University of Turku (Finland)
Abstract: The boosting estimation method is applied to investigate to what extent and at what horizons macroeconomic time series have nonlinear predictability coming from their own history. The results indicate that the U.S. macroeconomic time series have more exploitable nonlinear predictability than previous studies have found. On average, the most favorable out-of-sample performance is obtained by a two-stage procedure, where a conventional linear prediction model is fine-tuned by the boosting technique.