Tests for Normality in Linear Panel-data Models

Abstract

We propose a new command, xtsktest, for explaining nonnormalities in linear panel-data models. The command performs tests to explore skewness and excess kurtosis, allowing researchers to identify departures from Gaussianity in both error components of a standard panel regression, separately or jointly. The tests are based on recent results by Galvao et al. (2013, Journal of Multivariate Analysis 122: 35–52) and extend the classical Jarque–Bera normality test for the case of panel data.