Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/82428 
Year of Publication: 
2005
Series/Report no.: 
Sveriges Riksbank Working Paper Series No. 191
Publisher: 
Sveriges Riksbank, Stockholm
Abstract: 
We extend the standard approach to Bayesian forecast combination by forming the weights for the model averaged forecast from the predictive likelihood rather than the standard marginal likelihood. The use of predictive measures of fit offers greater protection against in-sample overfitting and improves forecast performance. For the predictive likelihood we show analytically that the forecast weights have good large and small sample properties. This is confirmed in a simulation study and an application to forecasts of the Swedish inflation rate where forecast combination using the predictive likelihood outperforms standard Bayesian model averaging using the marginal likelihood.
Subjects: 
Bayesian model averaging
Predictive likelihood
Partial Bayes factor
Training sample
Inflation rate
JEL: 
C11
C51
C52
Document Type: 
Working Paper

Files in This Item:
File
Size
884.9 kB





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