EmOptions: Display the current Expectation and Maximization options.

In ClustMMDD: Variable Selection in Clustering by Mixture Models for Discrete Data

Description

Display the Expectation and Maximization algorithm current options.

Usage

EmOptions()

Value

A list of EM options :

• epsi : The upper bound of the relative increasing on log-likelihood.

• nberSmallEM : The number of random parameter points from which to run small EMs. The estimated parameter point associated to the higher maximum log-likelihood is then used to initialise the final EM run.

• nberIterations : The number of iterations in each small EM.

• typeSmallEM : 0 = classic EM, 1 = SEM and 2 = CEM.

• typeEM : 0 = classic EM, 1 = SEM and 2 = CEM.

• nberMaxIterations : The maximum number of iterations in the final EM if the convergence is slow.

• putThreshold : The indication of whether all parameter estimates are positive.

Author(s)

Wilson Toussile.

References

• Dominique Bontemps and Wilson Toussile (2013) : Clustering and variable selection for categorical multivariate data. Electronic Journal of Statistics, Volume 7, 2344-2371, ISSN.

• Wilson Toussile and Elisabeth Gassiat (2009) : Variable selection in model-based clustering using multilocus genotype data. Adv Data Anal Classif, Vol 3, number 2, 109-134.

See Also

setEmOptions for setting EM options.

Examples

EmOptions()
setEmOptions(list(epsi = 1e-6))
EmOptions()
setEmOptions()	# To set default values
EmOptions()

ClustMMDD documentation built on May 2, 2019, 2:44 p.m.