B1557
Title: Assesing skewness, kurtosis and normality in linear mixed models
Authors: Alexandra Soberon - Universidad de Cantabria (Spain) [presenting]
Winfried Stute - University of Giessen (Germany)
Abstract: Linear mixed models provide a useful tool to fit continuous longitudinal data, and the random effects and error term are commonly assumed to have normal distribution. However, this restrictive assumption can result in a lack of robustness and needs to be tested. We propose a very simple tests for skewness, kurtosis and normality based on moment conditions of generalized least squares (GLS) residuals. To do it, estimating higher order momens is necessary and an alternative estimation procedure is developed. Regadless to other procedures in the literature, this is a simpler method that provides a closed-form expression even for the third and fourth order moments. In addition, no further distributional assumptions on neither random effects or error term are needed to show the consistency of the proposed estimators and tests statistics. Their finite sample performance is examined in a Monte Carlo study and the methodology is used to examine changes in the life expectancy and maternal and infant mortality rate of a sample of OECD countries.
Journal