Summarizing discriminant analysis based on Gaussian finite mixture modeling

Summary method for class "MclustDA".

Usage

# S3 method for MclustDA
summary(object, parameters = FALSE, newdata, newclass, ...)
# S3 method for summary.MclustDA
print(x, digits = getOption("digits"), ...)

Arguments

object

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

x

An object of class 'summary.MclustDA', usually, a result of a call to summary.MclustDA.

parameters

Logical; if TRUE, the parameters of mixture components are printed.

newdata

A data frame or matrix giving the test data.

newclass

A vector giving the class labels for the observations in the test data.

digits

The number of significant digits to use when printing.

...

Further arguments passed to or from other methods.

Value

The function summary.MclustDA computes and returns a list of summary statistics of the estimated MclustDA or EDDA model for classification.

Author

Luca Scrucca

See also

MclustDA, plot.MclustDA.

Examples

mod = MclustDA(data = iris[,1:4], class = iris$Species)
summary(mod)
#> ------------------------------------------------ 
#> 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 
summary(mod, parameters = TRUE)
#> ------------------------------------------------ 
#> 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
#> 
#> Class prior probabilities:
#>     setosa versicolor  virginica 
#>  0.3333333  0.3333333  0.3333333 
#> 
#> Class = setosa
#> 
#> Mixing probabilities: 0.8081091 0.1918909 
#> 
#> Means:
#>                   [,1]      [,2]
#> Sepal.Length 4.9484622 5.2483085
#> Sepal.Width  3.3627745 3.7026840
#> Petal.Length 1.4322810 1.5871556
#> Petal.Width  0.2036194 0.4244774
#> 
#> Variances:
#> [,,1]
#>               Sepal.Length  Sepal.Width Petal.Length   Petal.Width
#> Sepal.Length  0.1078221133  0.081427298 0.0088268249 -0.0001451869
#> Sepal.Width   0.0814272976  0.122899586 0.0033006551 -0.0025292826
#> Petal.Length  0.0088268249  0.003300655 0.0258364946  0.0006438246
#> Petal.Width  -0.0001451869 -0.002529283 0.0006438246  0.0033200166
#> [,,2]
#>               Sepal.Length  Sepal.Width Petal.Length   Petal.Width
#> Sepal.Length  0.1078221133  0.081427298 0.0088268249 -0.0001451869
#> Sepal.Width   0.0814272976  0.122899586 0.0033006551 -0.0025292826
#> Petal.Length  0.0088268249  0.003300655 0.0258364946  0.0006438246
#> Petal.Width  -0.0001451869 -0.002529283 0.0006438246  0.0033200166
#> 
#> Class = versicolor
#> 
#> Mixing probabilities: 1 
#> 
#> Means:
#>               [,1]
#> Sepal.Length 5.936
#> Sepal.Width  2.770
#> Petal.Length 4.260
#> Petal.Width  1.326
#> 
#> Variances:
#> [,,1]
#>              Sepal.Length Sepal.Width Petal.Length Petal.Width
#> Sepal.Length     0.261104     0.08348      0.17924    0.054664
#> Sepal.Width      0.083480     0.09650      0.08100    0.040380
#> Petal.Length     0.179240     0.08100      0.21640    0.071640
#> Petal.Width      0.054664     0.04038      0.07164    0.038324
#> 
#> Class = virginica
#> 
#> Mixing probabilities: 1 
#> 
#> Means:
#>               [,1]
#> Sepal.Length 6.588
#> Sepal.Width  2.974
#> Petal.Length 5.552
#> Petal.Width  2.026
#> 
#> Variances:
#> [,,1]
#>              Sepal.Length Sepal.Width Petal.Length Petal.Width
#> Sepal.Length     0.396256    0.091888     0.297224    0.048112
#> Sepal.Width      0.091888    0.101924     0.069952    0.046676
#> Petal.Length     0.297224    0.069952     0.298496    0.047848
#> Petal.Width      0.048112    0.046676     0.047848    0.073924
#> 
#> 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