Robust quadratic discriminant analysis is performed by replacing the classical group means and withing group covariance matrices by their robust equivalents.
Objects can be created by calls of the form
new("QdaCov", ...) but the
usual way of creating
QdaCov objects is a call to the function
QdaCov which serves as a constructor.
The (matched) function call.
Prior probabilities used, default to group proportions
number of observations in each class
the group means
the group covariance matrices
the inverse of the group covariance matrices
the determinants of the group covariance matrices
a character string giving the estimation method used
the training data set (same as the input parameter x of the constructor function)
grouping variable: a factor specifying the class for each observation.
Object of class
"CovControl" specifying which estimate to
use for the group means and covariances
"Qda", by class "QdaRobust", distance 2.
No methods defined with class "QdaCov" in the signature.
Valentin Todorov email@example.com
Todorov V & Filzmoser P (2009), An Object Oriented Framework for Robust Multivariate Analysis. Journal of Statistical Software, 32(3), 1–47. URL http://www.jstatsoft.org/v32/i03/.