Flexible Clustering for High-Dimensional Data via Mixtures of Joint Generalized Hyperbolic Models
arXiv:1705.03130 · doi:10.1002/sta4.177
Abstract
A mixture of joint generalized hyperbolic distributions (MJGHD) is introduced for asymmetric clustering for high-dimensional data. The MJGHD approach takes into account the cluster-specific subspace, thereby limiting the number of parameters to estimate while also facilitating visualization of results. Identifiability is discussed, and a multi-cycle ECM algorithm is outlined for parameter estimation. The MJGHD approach is illustrated on two real data sets, where the Bayesian information criterion is used for model selection.
References in corpus (7)
- High-Dimensional Data Clustering
- A Mixture of Generalized Hyperbolic Distributions
- Mixtures of Skew-t Factor Analyzers
- Estimating Common Principal Components in High Dimensions
- Clustering, Classification, Discriminant Analysis, and Dimension Reduction via Generalized Hyperbolic Mixtures
- Mixtures of Common Skew-t Factor Analyzers
- A Mixture of Generalized Hyperbolic Factor Analyzers