Multivariate Models to Forecast Portfolio Value at Risk: from the Dot- Com crisis to the global financial crisis

Vítor Manuel de Sousa Gabriel


This study analyzed market risk of an international
investment portfolio by means of a new
methodological proposal based on Value-at-
Risk, using the covariance matrix of multivariate
GARCH-type models and the extreme value
theory to realize if an international diversification
strategy minimizes market risk, and to determine
if the VaR methodology adequately captures
market risk, by applying Backtesting tests. To
this end, we considered twelve international
stock indexes, accounting for about 62% of the
world stock market capitalization, and chose the
period from the Dot-Com crisis to the current
global financial crisis. Results show that the
proposed methodology is a good alternative to
accommodate the high market turbulence and
can be considered as an adequate portfolio risk
management instrument.


Stock markets. Value at Risk. Multivariate GARCH models. Extreme value theory. Backtesting.


Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Av. da Liberdade, 532 - São Paulo - SP Brasil Cep. 01502-001 Tel. 55 11 3272-2340
 The RBGN adopts the Creative Commons licens CC-BY Creative Commons Attribution 4.0 

The RBGN has the financial support from Fundação Escola de Comércio Álvares Penteado - FECAP and development agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – CAPES.

Desenvolvido por:

Logomarca da Lepidus Tecnologia