In predicting conditional covariance matrices of financial portfolios, practitioners arerequired to choose among several alternative options, facing a number of differentsources of uncertainty. A first source is related to the frequency at which prices areobserved, either daily or intradaily. Using prices sampled at higher frequency inevitablyposes additional sources of uncertainty related to the selection of the optimal intradailysampling frequency and to the construction of the best realized estimator. Likewise,the choices of model structure and estimation method also have a critical role. Inorder to alleviate the impact of these sources of uncertainty, we propose a forecastcombination strategy based on the Model Confidence Set [MCS] to adaptively identifythe set of most accurate predictors. The combined predictor is shown to achievesuperior performance with respect to the whole model universe plus three additionalcompetitors, independently of the MCS or portfolio settings.
A Model Confidence Set approach to the combination of multivariate volatility forecasts
Alessandra Amendola;Giuseppe Storti;Vincenzo Candila;
2020-01-01
Abstract
In predicting conditional covariance matrices of financial portfolios, practitioners arerequired to choose among several alternative options, facing a number of differentsources of uncertainty. A first source is related to the frequency at which prices areobserved, either daily or intradaily. Using prices sampled at higher frequency inevitablyposes additional sources of uncertainty related to the selection of the optimal intradailysampling frequency and to the construction of the best realized estimator. Likewise,the choices of model structure and estimation method also have a critical role. Inorder to alleviate the impact of these sources of uncertainty, we propose a forecastcombination strategy based on the Model Confidence Set [MCS] to adaptively identifythe set of most accurate predictors. The combined predictor is shown to achievesuperior performance with respect to the whole model universe plus three additionalcompetitors, independently of the MCS or portfolio settings.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.