Document Search
Add To My Bookshelf Sign in or Register Save And Annotate
Keywords:

copula sign in to follow this
portfolio sign in to follow this
Student-t sign in to follow this
GARCH sign in to follow this
Normal mixture sign in to follow this
Categories:

VaR Methods sign in to follow this
--Evaluation/Comparison sign in to follow this
Half-Life:
Impact:
Discuss This Paper
Sign in to follow this page
Recent Comments
  more
Dynamic Forecasting of Portfolio Risk
Year Of Publication: 2005
Month Of Publication: December
Pages: 43
Download Count: 4
View Count: 1520
Comment Num: 0
Language: English
Source: working paper
Who Can Read: Free
Date: 7-4-2010
Publisher: Administrator
Summary
Di?erent univariate risk models are presented and their out-of-sample
Value-at-Risk (VaR) forecasting performance is back-tested on seventy-four US stocks.
The considered models feature two types of volatility forecasting techniques: static
and dynamic. The static (unconditional) models feature normal, Student-t and
two-component Normal Mixture (NM(2)) distributions. The back-tested models
incorporate dynamic features by means of Generalized Auto-Regressive Conditional
Heteroskedastic (GARCH) approach where the innovations follow normal, Student-
t and NM(2) distributions. The dynamic GARCH-type implementations based on
leptokurtic distributions signi¯cantly outperform all other realizations. To arrive
at portfolio VaR models, the univariate margins are combined in multivariate
distributions by means of elliptical copulas. For two real portfolios, it is demonstrated
that Student-t copulas with degrees of freedom 4-6 provide an excellent modelling tool
for obtaining joint d
Author(s)
Ivanov, Stanislav Sign in to follow this author
Jordan, Richard Sign in to follow this author
Panajotova, Biliana Sign in to follow this author
This document's citation network:
Similar Documents:
Close window
Sign up in one step, no personal information required. Already a Member?



Email:
Repeat Email:
User Name:
Password:
Confirm Password:

Sign Up


Welcome to GloriaMundi!
Thanks for singning up



continue or edit your profile