MixtureMissing

CRANMixtureMissingMaRDI QIDQ77647

Robust and Flexible Model-Based Clustering for Data Sets with Missing Values at Random

Hung Tong, Cristina Tortora

Last update: 17 January 2024

Copyright license: GNU General Public License, version 3.0, GNU General Public License, version 2.0

Software version identifier: 2.0.0, 1.0.0, 1.0.1, 1.0.2, 2.0.0, 3.0.0, 3.0.1

Description

Implementations of various robust and flexible model-based clustering methods for data sets with missing values at random. Two main models are: Multivariate Contaminated Normal Mixture (MCNM, Tong and Tortora, 2022, <doi:10.1007/s11634-021-00476-1>) and Multivariate Generalized Hyperbolic Mixture (MGHM, Wei et al., 2019, <doi:10.1016/j.csda.2018.08.016>). Mixtures via some special or limiting cases of the multivariate generalized hyperbolic distribution are also included: Normal-Inverse Gaussian, Symmetric Normal-Inverse Gaussian, Skew-Cauchy, Cauchy, Skew-t, Student's t, Normal, Symmetric Generalized Hyperbolic, Hyperbolic Univariate Marginals, Hyperbolic, and Symmetric Hyperbolic.

Cites work

• Model-based clustering and outlier detection with missing data
• Mixtures of generalized hyperbolic distributions and mixtures of skew-t distributions for model-based clustering with incomplete data