Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/68509
Authors: 
Costantini, Mauro
Kunst, Robert M.
Year of Publication: 
2011
Series/Report no.: 
Economics Series, Institute for Advanced Studies 276
Abstract: 
In evaluating prediction models, many researchers flank comparative ex-ante prediction experiments by significance tests on accuracy improvement, such as the Diebold-Mariano test. We argue that basing the choice of prediction models on such significance tests is problematic, as this practice may favor the null model, usually a simple benchmark. We explore the validity of this argument by extensive Monte Carlo simulations with linear (ARMA) and nonlinear (SETAR) generating processes. For many parameter constellations, we find that utilization of additional significance tests in selecting the forecasting model fails to improve predictive accuracy.
Subjects: 
forecasting
time series
predictive accuracy
model selection!
JEL: 
C22
C52
C53
Document Type: 
Working Paper

Files in This Item:
File
Size
244.22 kB





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