On the non‐identifiability of unified skew‐normal distributions
Abstract
We investigate the non‐identifiability of the multivariate unified skew‐normal distribution under permutation of its latent variables. We show that the non‐identifiability issue also holds with other parameterizations and extends to the family of unified skew‐elliptical distributions and more generally to selection distributions. We provide several suggestions to make the unified skew‐normal model identifiable and describe various sub‐models that are identifiable.Keywords
All Related Versions
Funding Information
- King Abdullah University of Science and Technology
This publication has 14 references indexed in Scilit:
- Conjugate Bayes for probit regression via unified skew-normal distributionsBiometrika, 2019
- A skew Gaussian decomposable graphical modelJournal of Multivariate Analysis, 2016
- On Some Properties of the Unified Skew Normal DistributionJournal of Statistical Theory and Practice, 2013
- Non-Gaussian modeling of spatial data using scale mixing of a unified skew Gaussian processJournal of Multivariate Analysis, 2013
- Estimation of Parameters of the Unified Skew Normal Distribution Using the Method of Weighted MomentsJournal of Statistical Theory and Practice, 2012
- A unified view on skewed distributions arising from selectionsCanadian Journal of Statistics, 2006
- On the Unification of Families of Skew‐normal DistributionsScandinavian Journal of Statistics, 2006
- On fundamental skew distributionsJournal of Multivariate Analysis, 2005
- The Closed Skew-Normal Distribution2004
- A Bayesian interpretation of the multivariate skew-normal distributionStatistics & Probability Letters, 2003