klaR (version 0.6-2)

predict.pvs: predict method for pvs objects

Description

Prediction of class membership and posterior probabilities using pairwise variable selection.

Usage

## S3 method for class 'pvs':
predict(object, newdata, quick = FALSE, detail = FALSE, ...)

Arguments

object

an object of class pvs, as that created by the function pvs

newdata

a data frame or matrix containing new data. If not given the same datas as used for training the pvs-model are used.

quick

indicator (logical), whether a quick, but less accurate computation of posterior probabalities should be used or not.

detail

indicator (logical), whether the returned object includes additional information about the posterior probabilities for each date in each submodel.

...

Further arguments are passed to underlying predict calls.

Value

• a list with components:
• class
  the predicted classes
• posterior
  posterior probabilities for the classes
• details
  (only if details=TRUE. A list containing matrices of posterior probabalities computated by the classification method for each case and classpair.

concept

Pairwise variable selection for classification

Details

If quick=FALSE the posterior probabilites for each case are computed using the pairwise coupling algorithm presented by Hastie, Tibshirani (1998). If quick=FALSE a much quicker solution is used, which leads to less accurate posterior probabalities. In almost all cases it doesn't has a negative effect on the classification result.

References

Szepannek, G. and Weihs, C. (2006) Variable Selection for Classification of More than Two Classes Where the Data are Sparse. In From Data and Information Analysis to Kwnowledge Engineering., eds Spiliopolou, M., Kruse, R., Borgelt, C., Nuernberger, A. and Gaul, W. pp. 700-708. Springer, Heidelberg.

See Also

pvs

Examples

## learn a pvs-model with half of the Satellite dataset, 
## using "ks.test" as selection and "qda" as classification method:

library("mlbench")
data("Satellite")

model <- pvs(classes ~ ., Satellite[1:3218,], method = "qda", vs.method = "ks.test")
model # short summary, showing the class-pairs of the submodels and the selected variables

## now predict on the rest of the data set:

## pred <- predict(model,Satellite[3219:6435,]) # takes some time
pred <- predict(model,Satellite[3219:6435,], quick=TRUE) # that's much quicker

## now you can look at the predicted classes:
pred$class 
## or the posterior probabilities:
pred$posterior