Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/233339 
Year of Publication: 
2018
Series/Report no.: 
Discussion paper No. 124
Publisher: 
Aboa Centre for Economics (ACE), Turku
Abstract: 
We apply the boosting estimation method to investigate to what ex-tent and at what horizons macroeconomic time series have nonlinearpredictability coming from their own history. Our results indicate thatthe U.S. macroeconomic time series have more exploitable nonlinearpredictability than previous studies have found. On average, the mostfavorable out-of-sample performance is obtained by a two-stage proce-dure, where a conventional linear prediction model is fine-tuned by theboosting technique.
Subjects: 
boosting
forecasting
linear autoregression
mean squarederror
non-linearity
JEL: 
C22
C53
E27
E37
E47
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.