Aitken: Aitken Acceleration
In upclass: Updated Classification Methods using Unlabeled Data

Description Usage Arguments Details Value References See Also Examples

Calculates the Aitken acceleration estimate of the final converged maximized log-likelihood.

1	Aitken(ll)

`ll`	A vector of three consecutive log-likelihood values

The final converged maximized log-likelihood can be used to determine convergence, i.e., for use in determining convergence of the EM algorithm. It can be used alongside other functions in the upclass package to decide whether or not the log-likelihood has converged.

The return value as a list with the following components:

`ll`	The most current estimate for the log-likelihood
`linf`	An estimate of the final converged maximized log-likelihood
`a`	The Aitken acceleration value where `0 <= a <= 1`

Boehning, D., Dietz, E., Schaub, R., Schlattmann, and Lindsay, B. (1994, June). The distribution of the likelihood ratio for mixtures of densities from the one-parameter exponential family. Annals of the Institute of Statistical Mathematics 46 (2), 373-388.

upclassifymodel, noupclassifymodel.

1 2	ll<-c(-261, -257.46,-256.4) Aitken(ll)

Loading required package: mclust
Package 'mclust' version 5.3
Type 'citation("mclust")' for citing this R package in publications.
$ll
[1] -256.4

$linf
[1] -254.8869

$a
[1] 0.299435