EcoSta 2019: Start Registration
View Submission - EcoSta2019
A0159
Title: Recent advances on mixtures of skew distributions for modelling heterogeneous and asymmetric data Authors:  Sharon Lee - University of Queensland (Australia)
Geoffrey McLachlan - University of Queensland (Australia) [presenting]
Abstract: Finite mixtures of skew distributions provide a flexible tool for modelling heterogeneous data with asymmetric distributional features. In recent years, several skew variants of the multivariate normal and t-distributions have been proposed. However, attention has been focused mainly on distributions that are limited to modelling skewness concentrated in a single direction in the feature sample space. We consider a general class of skew distributions that can model various types of skewness and asymmetry in the data, including being able to accommodate multiple directions of skewness. We also consider mixtures of skew factor analyzers for applications to high-dimensional data. The usefulness and potential of the proposed models are demonstrated using real datasets.