Overfitting Bayesian Mixtures of Factor Analyzers with Parsimonious Covariance and Unknown Number of Components
Gibbs sampling for
Complete log-likelihood function for xCx models.
Complete log-likelihood function for xUx models and
Complete log-likelihood function for xCx models and
Complete log-likelihood function for xUx models.
Computation and simulations
Computation and simulations for CCU
Computation and simulations for CUU
Computation and simulations for .
Computation and simulations for .
Computation and simulations
Compute sufficient statistics
Compute sufficient statistics given mu
Compute sufficient statistics for
Compute quantiles for the correlation matrix.
Compute quantiles for the covariance matrix.
Apply label switching algorithms
tools:::Rd_package_title("fabMix")
Main function
Function to estimate the CUC and CCC models
Function to estimate the CCU and CUU models
Function to estimate the UUU or UCU models in case of missing values
Function for model-level parallelization
Function to estimate the UUC and UCC models
Function to estimate the UUU and UCU model
Compute information criteria
Log-density function of the Dirichlet distribution
Simulate from the Dirichlet distribution
Log-likelihood of the mixture model
Log-likelihood of the mixture model for and same variance of err...
Log-likelihood of the mixture model
Log-likelihood of the mixture model for
MCMC sampler for
MCMC sampler for and same error variance parameterization
Basic MCMC sampler for the UCU model
Basic MCMC sampler for the CCC model
Basic MCMC sampler for the CCU model
Basic MCMC sampler for the CUC model
Basic MCMC sampler for the CUU model
Basic MCMC sampler for the case of missing data
Basic MCMC sampler for the UUU model
Basic MCMC sampler for the case of missing data and different error va...
Basic MCMC sampler for the UCC model
Basic MCMC sampler for the UUC model
Plot function
Print function
Read Lambda values.
Synthetic data generator
Synthetic data generator 2
Summary method
Gibbs sampling for in xCx model
Gibbs sampling for in xUx model
Gibbs sampling for
Gibbs sampling for for Cxx model
Gibbs sampling for
Gibbs sampling for per component for
Gibbs sampling for per component for
Gibbs sampling for per component
Gibbs sampling for for xCC models
Gibbs sampling for per component for xUC models
Gibbs sampling for
Gibbs sampling for for
Gibbs sampling for for
Collapsed Gibbs for using matrix inversion lemma
Collapsed Gibbs for using matrix inversion lemma
Collapsed Gibbs for
Collapsed Gibbs for
Model-based clustering of multivariate continuous data using Bayesian mixtures of factor analyzers (Papastamoulis (2019) <DOI:10.1007/s11222-019-09891-z> (2018) <DOI:10.1016/j.csda.2018.03.007>). The number of clusters is estimated using overfitting mixture models (Rousseau and Mengersen (2011) <DOI:10.1111/j.1467-9868.2011.00781.x>): suitable prior assumptions ensure that asymptotically the extra components will have zero posterior weight, therefore, the inference is based on the ``alive'' components. A Gibbs sampler is implemented in order to (approximately) sample from the posterior distribution of the overfitting mixture. A prior parallel tempering scheme is also available, which allows to run multiple parallel chains with different prior distributions on the mixture weights. These chains run in parallel and can swap states using a Metropolis-Hastings move. Eight different parameterizations give rise to parsimonious representations of the covariance per cluster (following Mc Nicholas and Murphy (2008) <DOI:10.1007/s11222-008-9056-0>). The model parameterization and number of factors is selected according to the Bayesian Information Criterion. Identifiability issues related to label switching are dealt by post-processing the simulated output with the Equivalence Classes Representatives algorithm (Papastamoulis and Iliopoulos (2010) <DOI:10.1198/jcgs.2010.09008>, Papastamoulis (2016) <DOI:10.18637/jss.v069.c01>).