predict.epx {EPX}R Documentation

Predict with an "epx" object

Description

Predicted values based on an "epx" object; may specify different base classifier arguments than those used for phalanx-formation.

Usage

## S3 method for class 'epx'
predict(object, newdata, classifier.args = list(), ...)

Arguments

object

Object of class "epx".

newdata

An optional data frame specifiying variables with which to predict; if omitted and classifier.args are not specified, the fitted (ensembled) values are used.

classifier.args

Additional arguments for the base classifier; same base classifier as that used for phalanx-formation (specified in epx).

...

Further arguments passed to or from other methods.

Value

Numeric vector of predicted values (double).

Examples

# Example with data(harvest)

## Phalanx-formation using a base classifier with 50 trees (default = 500)

set.seed(761)
model <- epx(x = harvest[, -4], y = harvest[, 4],
             classifier.args = list(ntree = 50))

## Predict training values without additional classifier.args and newdata
## returns the object's ENSEMBLED.FITS
all.equal(predict(model), model$ENSEMBLED.FITS)

## Predict training values using 100 trees (default = 500)
set.seed(761)
preds100 <- predict(model, classifier.args = list(ntree = 100))

## Predict test values by passing dataframe of test predictors to newdata as
## with the predict(model, newdata = . ) function etc.


[Package EPX version 1.0.4 Index]