 |
Forecasting Realized Volatility Measures with Multivariate and Univariate Models: The Case of The US Banking Sector
1 : Maastricht University
(UM)
P.O.Box 616, 6200 MD Maastricht -
Netherlands
2 : Maastricht University
3 : Università di Roma Tor Vergata
4 : Interactive Data Kler'S
* : Corresponding author
This paper aims at evaluating the forecasting performances of a set of univariate fractional white noise processes versus multivariate factor models for realized volatility measures. We do not provide a new horse race comparaison of different approaches. Instead, through forecast accuracy analyses, we wish to evaluate the underlying mechanisms that has generated realized volatilities of 13 banking sector asset returns.
