Elliptical multivariate semiparametric linear regression model with equicorrelated random errors: some theoretical aspects of the estimation process and diagnostic analysis

Abstract

In this thesis, we will study elliptical multivariate semiparametric linear regression model with equicorrelated random errors, specifically, some theoretical aspects of the estimation process and diagnostic analysis. An estimation method based in penalized likelihood function is developed. In addition, the local influence curvature was developed to assess the sensitivity of the estimators to the observations that are influential.