Events and Meetings of Italian Statistical Society, Advances in Latent Variables - Methods, Models and Applications

Font Size: 
A thick modeling approach to multivariate volatility prediction
Alessandra Amendola, Giuseppe Storti

Last modified: 2013-06-16


This paper proposes a modified approach to the combination of forecasts from multivariate volatility models where the combination is performed over the subset of the best performing models. Such a subset is identified over a rolling window using the Model Confidence Set approach. An application to a vast dimensional portfolio of 50 US stocks shows that i) in non-extreme volatility periods the use of forecast combinations allows to improve over the predictive accuracy of the single candidate models ii) performing the combination over the subset of most accurate models does not significantly reduce the accuracy of the combined predictor.

Full Text: PDF