Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/152628 
Year of Publication: 
2002
Series/Report no.: 
ECB Working Paper No. 194
Publisher: 
European Central Bank (ECB), Frankfurt a. M.
Abstract: 
The extension of GARCH models to the multivariate setting has been fraught with difficulties. In this paper, we suggest to work with univariate portfolio GARCH models. We show how the multivariate dimension of the portfolio allocation problem may be recovered from the univariate approach. The main tool we use is the "variance sensitivity analysis", which measures the change in the portfolio variance as a consequence of an infinitesimal change in the portfolio allocation. We derive the sensitivity of the univariate portfolio GARCH variance to the portfolio weights, by analytically computing the derivatives of the estimated GARCH variance with respect to these weights. We suggest a new and simple method to estimate full variance-covariance matrices of portfolio assets. An application to real data portfolios shows how to implement our methodology and compares its performance against that of selected popular alternatives.
Subjects: 
Dynamic Correlations
GARCH
risk management
Sensitivity Analysis
JEL: 
C32
C53
G15
Document Type: 
Working Paper

Files in This Item:
File
Size
387.45 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.