TY - JOUR
T1 - Scale and shape mixtures of multivariate skew-normal distributions
AU - Arellano-Valle, Reinaldo B.
AU - Ferreira, Clécio S.
AU - Genton, Marc G.
N1 - Funding Information:
This research was supported by Fondecyt (Chile) 1120121 and 1150325 , by FAPEMIG (Minas Gerais State Foundation for Research Development) grant CEX APQ 01944/17 , and by the King Abdullah University of Science and Technology (KAUST) . We thank the Editor, Associate Editor and four anonymous reviewers for comments that improved the paper. We also thank Prof. Adelchi Azzalini for suggesting Proposition 1 during a seminar presentation of this work at the University of Padova and Prof. Mauricio Castro for some initial discussions on the topic of this paper.
Publisher Copyright:
© 2018 Elsevier Inc.
PY - 2018/7
Y1 - 2018/7
N2 - We introduce a broad and flexible class of multivariate distributions obtained by both scale and shape mixtures of multivariate skew-normal distributions. We present the probabilistic properties of this family of distributions in detail and lay down the theoretical foundations for subsequent inference with this model. In particular, we study linear transformations, marginal distributions, selection representations, stochastic representations and hierarchical representations. We also describe an EM-type algorithm for maximum likelihood estimation of the parameters of the model and demonstrate its implementation on a wind dataset. Our family of multivariate distributions unifies and extends many existing models of the literature that can be seen as submodels of our proposal.
AB - We introduce a broad and flexible class of multivariate distributions obtained by both scale and shape mixtures of multivariate skew-normal distributions. We present the probabilistic properties of this family of distributions in detail and lay down the theoretical foundations for subsequent inference with this model. In particular, we study linear transformations, marginal distributions, selection representations, stochastic representations and hierarchical representations. We also describe an EM-type algorithm for maximum likelihood estimation of the parameters of the model and demonstrate its implementation on a wind dataset. Our family of multivariate distributions unifies and extends many existing models of the literature that can be seen as submodels of our proposal.
KW - EM-algorithm
KW - Scale mixtures of normal distributions
KW - Scale mixtures of skew-normal distributions
KW - Shape mixtures of skew-normal distributions
KW - Skew scale mixtures of normal distributions
KW - Skew-normal distribution
UR - http://www.scopus.com/inward/record.url?scp=85043577337&partnerID=8YFLogxK
U2 - 10.1016/j.jmva.2018.02.007
DO - 10.1016/j.jmva.2018.02.007
M3 - Article
AN - SCOPUS:85043577337
SN - 0047-259X
VL - 166
SP - 98
EP - 110
JO - Journal of Multivariate Analysis
JF - Journal of Multivariate Analysis
ER -