The orthogonal skew model: computationally efficient multivariate skew-normal and skew-t distributions with applications to model-based clustering

Abstract

We introduce a parameterization for the multivariate skew normal and skew-t distributions, which enforces an orthogonal structure on the skewness parameter. This approach provides substantial benefits in computational efficiency during parameter estimation, resulting in a model which strikes an excellent balance between flexibility and model-fitting feasibility. We illustrate this primarily through implementing the proposed distributions in a mixture model-based clustering framework. We compare to competing skew distributions via both simulated and real data analyses, reporting both computation time and model-fit metrics.