Forecast Combination and Model Averaging Using Predictive Measures

Jana Eklund and Sune Karlsson ()

Econometric Reviews, 2007, vol. 26, issue 2-4, pages 329-363

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 when uninformative priors on the model parameters are used and improves forecast performance. For the predictive likelihood we argue that the forecast weights have good large and small sample properties. This is confirmed in a simulation study and in 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.

Keywords: Bayesian model averaging; Inflation rate; Partial Bayes factor; Predictive likelihood; Training sample; Uninformative priors (search for similar items in EconPapers)
View citations in EconPapers

Downloads: (external link)
http://www.informawo ... 40C6AD35DC6213A474B5 (text/html)
Access to full text is restricted to subscribers.

Related works:
Working Paper: Forecast Combination and Model Averaging using Predictive Measures (2005)
Working Paper: Forecast Combination and Model Averaging Using Predictive Measures (2005)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Ordering information: This journal article can be ordered from
http://www.tandf.co.uk/journals/subscription.html

Access Statistics for this article

Econometric Reviews is edited by Esfandiar Maasoumi

More articles in Econometric Reviews from Taylor and Francis Journals
Series data maintained by Christopher F. Baum ().