Resumen

NUNEZ MORA, José Antonio  y  MATA MATA, Leovardo. Covariances matrix under the multivariate-Gh funtion to desing portfolios. Contad. Adm [online]. 2016, vol.61, n.3, pp.535-550. ISSN 0186-1042.  https://doi.org/10.1016/j.cya.2015.11.009.

In this paper we developed the estimation implementation of the generalized hyperbolic multivariate (GH) distribution with a non-fixed Bessel function. The covariance matrix estimated through the GH distribution complements the use of the Markowitz procedure to construct an efficient portfolio and reduce the variation coefficient of the expected return. The data are from the Stockholm index 30 from January 2010 to April 2014.

Palabras llave : Expectation-maximization algorithm; Generalized hyperbolic distribution; Markowitz portfolio; Covariance matrix.