fd.est: Fitting a Flexible Dirichlet distribution

In Compositional: Compositional Data Analysis

Fitting a Flexible Dirichlet distribution

Description

Fitting a Flexible Dirichlet distribution.

Usage

fd.est(x, ini.iter = 50, final.iter = 100)

Arguments

x	A matrix or a dataframe containing the compositional data.
ini.iter	Number of iterations for the initial SEM step. Default value is 50.
final.iter	Number of iterations for the final EM step. Default value is 100.

Details

For more information see the references.

Value

A list including:

alpha	Estimated values of the parameter vector \alpha.
prob	Estimated values of the parameter vector p.
tau	Estimated value of the parameter tau.
loglik	The Log-likelihood value.

Author(s)

Michail Tsagris ported from the R package FlexDir. [e-mail].

References

Ongaro A. and Migliorati S. (2013). A generalization of the Dirichlet distribution. Journal of Multivariate Analysis, 114, 412–426.

Migliorati S., Ongaro A. and Monti G. S. (2017). A structured Dirichlet mixture model for compositional data: inferential and applicative issues. Statistics and Computing, 27, 963–983.

See Also

rfd, rfd

Examples

x <- rfd(n = 50, a = c(12, 11, 10), p = c(0.25, 0.25, 0.5), tau = 4 )
ela <- fd.est(x, ini.iter = 10, final.iter = 20)
ela

Compositional documentation built on Oct. 23, 2023, 5:09 p.m.