sda-package {sda}R Documentation

The sda package

Description

This package performs linear and diagonal discriminant analysis where the classifier is trained using Stein-type shrinkage estimators.

Specifically, three separate shrinkage estimators are employed for the

class frequencies
following Hausser and Strimmer(2008),
variances
following Opgen-Rhein and Strimmer (2007), and
correlations
following Sch"afer and Strimmer (2005).

The estimates are plugged into the LDA and DDA discriminant function for prediction. Note that the three corresponding regularization parameters are obtained analytically without resorting to computer intensive resampling.

This approach is particularly suited for high-dimensional classification.

Author(s)

Korbinian Strimmer (http://strimmerlab.org/)

References

See website: http://strimmerlab.org/software/sda/

See Also

sda, predict.sda


[Package sda version 1.0.0 Index]