New Insights on the Multivariate Skew Exponential Power Distribution

ABSTRACT The multivariate exponential power is a useful distribution for modeling departures from normality in data by means of tail weight scalar parameter that regulates the non-normality model. incorporation shape asymmetry vector into model serves to account potential asymmetries and gives rise skew distribution. This work aimed at revisiting taking as starting point its formulation scale mixture skew-normal distributions. paper provides some highlights theoretical insights on role played parameters assess two complementary aspects such directional behavior regardless asymmetry. intuition behind both issues relies well-known mathematical ideas about skewness maximization convex transform stochastic orderings.