Discussion Paper Details

Please find the details for DP5268 in an easy to copy and paste format below:

Full Details   |   Bibliographic Reference

Full Details

Title: Forecast Combination and Model Averaging Using Predictive Measures

Author(s): Jana Eklund and Sune Karlsson

Publication Date: October 2005

Keyword(s): Bayesian model averaging, inflation rate, partial Bayes factor, predictive likelihood and training sample

Programme Area(s): International Macroeconomics

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.

For full details and related downloads, please visit:

Bibliographic Reference

Eklund, J and Karlsson, S. 2005. 'Forecast Combination and Model Averaging Using Predictive Measures'. London, Centre for Economic Policy Research.