Optimal prediction pools
We consider the properties of weighted linear combinations of prediction models, or linear pools, evaluated using the log predictive scoring rule. Although exactly one model has limiting posterior probability, an optimal linear combination typically includes several models with positive weights. We derive several interesting results: for example, a model with positive weight in a pool may have zero weight if some other models are deleted from that pool. The results are illustrated using S&P 500 returns with six prediction models. In this example models that are clearly inferior by the usual scoring criteria have positive weights in optimal linear pools.
Year of publication: |
2011
|
---|---|
Authors: | Geweke, John ; Amisano, Gianni |
Published in: |
Journal of Econometrics. - Elsevier, ISSN 0304-4076. - Vol. 164.2011, 1, p. 130-141
|
Publisher: |
Elsevier |
Keywords: | Forecasting Log scoring Model combination S&P 500 returns |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Hierarchical Markov normal mixture models with applications to financial asset returns
Geweke, John, (2011)
-
Prediction using several macroeconomic models
Amisano, Gianni, (2013)
-
Analysis of Variance for Bayesian Inference
Geweke, John, (2014)
- More ...