Log-Likelihood of a MclustDA object

Returns the log-likelihood for a MclustDA object.

Usage

# S3 method for MclustDA
logLik(object, data, ...)

Arguments

object

an object of class 'MclustDA' resulting from a call to MclustDA.

data

the data for which the log-likelihood must be computed. If missing, the observed data from the 'MclustDA' object is used.

...

further arguments passed to or from other methods.

Value

Returns an object of class 'logLik' with an element providing the maximized log-likelihood, and further arguments giving the number of (estimated) parameters in the model ("df") and the sample size ("nobs").

Author

Luca Scrucca

See also

MclustDA.

Examples

# \donttest{
irisMclustDA <- MclustDA(iris[,1:4], iris$Species)
summary(irisMclustDA)
#> ------------------------------------------------ 
#> Gaussian finite mixture model for classification 
#> ------------------------------------------------ 
#> 
#> MclustDA model summary: 
#> 
#>  log-likelihood   n df       BIC
#>       -172.8135 150 47 -581.1269
#>             
#> Classes       n     % Model G
#>   setosa     50 33.33   EEE 2
#>   versicolor 50 33.33   XXX 1
#>   virginica  50 33.33   XXX 1
#> 
#> Training confusion matrix:
#>             Predicted
#> Class        setosa versicolor virginica
#>   setosa         50          0         0
#>   versicolor      0         48         2
#>   virginica       0          1        49
#> Classification error = 0.02 
#> Brier score          = 0.0116 
logLik(irisMclustDA)
#> 'log Lik.' -172.8135 (df=47)
# }