Optimal portfolio allocation decisions based on a copula-adcc-garch approach

• María del Mar Miralles Quirós ^[1] ; Jose Luis Miralles Quirós ^[1] ; Luis Miguel Valente Gonçalves ^[2]
  1. [1] Universidad de Extremadura

     Universidad de Extremadura

     Badajoz, España

     
  2. [2] Universidade Autónoma de Lisboa

     Universidade Autónoma de Lisboa

     Socorro, Portugal

     
• Localización: XXX Jornadas Luso-Espanholas de Gestão Científica: cooperação transfronteiriça. Desenvolvimento e coesão territorial. Livro de resumos / Paula Fernandes (aut.), 2020, ISBN 978-972-745-273-6, pág. 216
• Idioma: inglés
• Enlaces
  • Texto Completo Libro
• Resumen

  • In order to find the optimal portfolio, the correct estimation of linkages among assets is fundamental. In this sense, we propose a Copula-based Multivariate GARCH approach which permits modelling time-varying conditional correlation as well as combine copulas and the Extreme Value Theory, which analyzes events that deviate sharply from the norm. We use the volatility forecasts from the proposed model to construct a global minimum variance portfolio which is evaluated in basis of two performance measures: the Sharpe ratio and the modified Sharpe ratio. Our overall results show that the proposed approach leads to a significant improvement on the out-of-sample portfolio performances when are compared to those obtained from the naïve rule and the use of a standard Multivariate GARCH specification. These results persist during periods of extreme events and with moderate transaction costs. Our findings have important implications for both academic and market practitioners.