mixturebf-class function

Class "mixturebf"

Stores the output of Bayesian model selection for mixture models, e.g. as produced by function bfnormmix.

Methods are provided for retrieving the posterior probability of a given number of mixture components, posterior means and posterior samples of the mixture model parameters. 1.1

class

Objects from the Class

Typically objects are automatically created by a call to bfnormmix.

Slots

The class has the following slots:

• postprob: data.frame containing posterior probabilities for different numbers of components (k) and log-posterior probability of a component being empty (contain no individuals)
• p: Number of variables in the data to which the model was fit
• n: Number of observations in the data to which the model was fit
• priorpars: Prior parameters used when fitting the model
• postpars: Posterior parameters for a 1-component mixture, e.g. for a Normal mixture the posterior is N(mu1,Sigma/prec) IW(nu1,S1)
• mcmc: For each considered value of k, posterior samples for the parameters of the k-component model are stored

Methods

• coef: Computes posterior means for all parameters
• show: signature(object = "mixturebf"): Displays general information about the object.
• postProb: signature(object = "mixturebf"): Extracts posterior model probabilities, Bayes factors and posterior probability of a cluster being empty
• postSamples: signature(object = "mixturebf"): Extracts posterior samples

References

Fuquene J., Steel M.F.J., Rossell D. On choosing mixture components via non-local priors. 2018. arXiv

Author(s)

David Rossell

See Also

See also bfnormmix

Examples

showClass("mixturebf")