Gaussian Parsimonious Clustering Models with Covariates and a Noise Component
Aitken Acceleration
Convert MoEClust objects to the Mclust class
Drop constant variables from a formula
Drop unused factor levels to predict from unseen data
Account for extra variability in covariance matrices with expert covar...
Compute the Frobenius (adjusted) Rand index
Force diagonal elements of a triangular matrix to be positive
Average posterior probabilities of a fitted MoEClust model
MoEClust: Gaussian Parsimonious Clustering Models with Covariates and ...
Choose the best MoEClust model
Set control values for use with MoEClust
MoEClust BIC, ICL, and AIC Model-Selection Criteria
C-step for MoEClust Models
Density for MoEClust Mixture Models
Entropy of a fitted MoEClust model
E-step for MoEClust Models
Generalised Pairs Plots for MoEClust Mixture Models
Mahalanobis Distance Outlier Detection for Multivariate Response
Show the NEWS file
Model Selection Criteria Plot for MoEClust Mixture Models
Plot MoEClust Gating Network
Plot the Log-Likelihood of a MoEClust Mixture Model
Plot the Similarity Matrix of a MoEClust Mixture Model
Stepwise model/variable selection for MoEClust models
Plot Clustering Uncertainties
MoEClust: Gaussian Parsimonious Clustering Models with Covariates and ...
Approximate Hypervolume Estimate
Plot MoEClust Results
Predictions from MoEClust expert networks
Predictions from MoEClust gating networks
Predictions for MoEClust models
Quantile-Based Clustering for Univariate Data
Clustering via parsimonious Gaussian Mixtures of Experts using the MoEClust models introduced by Murphy and Murphy (2020) <doi:10.1007/s11634-019-00373-8>. This package fits finite Gaussian mixture models with a formula interface for supplying gating and/or expert network covariates using a range of parsimonious covariance parameterisations from the GPCM family via the EM/CEM algorithm. Visualisation of the results of such models using generalised pairs plots and the inclusion of an additional noise component is also facilitated. A greedy forward stepwise search algorithm is provided for identifying the optimal model in terms of the number of components, the GPCM covariance parameterisation, and the subsets of gating/expert network covariates.