Proceedings of the 19th International Academic Conference, Florence

BEHAVIOR OF COVARIANCE MATRICES WITH EQUI-CORRELATION APPROACH

R. REYTIER, A. BLANES, Q. GAUCHER, S. THIAM, P. DEBLED

Abstract:

Funds and asset managers are increasingly concerned with quantitative and econometric model in order to apply their portfolio models. The main goal of this publication is to study the behavior and the proportions of a stock portfolio from CAC All-Tradable with these kinds of models and compare the results with the historical approach. A GARCH (1,1) process has been used for modelling each asset volatility and Engle dynamic equi-correlation model to forecast covariance matrices. From a small amount of underlying values, the question is raised whether forecasted covariance matrix is more relevant than traditional variance-covariance matrix in a context of minimum variance portfolio model.

Keywords: Volatility - Correlation – Equi-Correlation - GARCH (1,1) - Portfolio Selection - Asset Allocation- Covariance Matrix – Minimum Variance Portfolio.

DOI: 10.20472/IAC.2015.019.116

PDF: Download



Copyright © 2020 The International Institute of Social and Economic Sciences, www.iises.net