SparseLearner (version 1.0-2)

Predict.bagging: Make predictions for new data from a 'bagging' object.

Description

This function makes predictions for new data from a bagging LASSO linear or logistic regression model, using the stored 'bagging' object, with or without the use of trimmed bagging strategy.

Usage

Predict.bagging(object, newx, y = NULL, trimmed = FALSE, scale.trimmed = 0.75)

Arguments

object

a fitted 'bagging' object.

newx

matrix of new values for x at which predictions are to be made. Must be a matrix. See documentation for Bagging.lasso.

y

response variable. Defaults to NULL. If the response variable for the newx matrix is known and input, the corresponding validation measures can be calculated for evaluating prediction performance.

trimmed

logical. Should a trimmed bagging strategy be performed? Defaults to FALSE. This argument should correspond to the same setting in the Bagging.lasso function. See documentation for Bagging.lasso.

scale.trimmed

the portion to trim of the "worst" based-level models, in the sense of having the largest error rates, and to average only over the most accurate base-level models. Defaults to 0.75.

Value

y.new

the predicted values of response vector y.

probabilities

the predicted probabilities of response vector y.

predicted.matrix

the matrix of predicted values of response vector y based on the base-level LASSO regression models.

bagging.prediction

the performance of bagging prediction accordig to the model validation measures defined.

Details

This function makes a prediction based on the object fitted by the Bagging.lasso model.

References

[1] Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5-32.

[2] Croux, C., Joossens, K., & Lemmens, A. (2007). Trimmed bagging. Computational Statistics & Data Analysis, 52(1), 362-368.

Examples

library(mlbench)
set.seed(0123)
mydata <- mlbench.threenorm(100, d=10)
x <- mydata$x
y <- mydata$classes
mydata <- as.data.frame(cbind(x, y))
colnames(mydata) <- c(paste("A", 1:10, sep=""), "y")
mydata$y <- ifelse(mydata$y==1, 0, 1)
# Split into training and testing data.
S1 <- as.vector(which(mydata$y==0))
S2 <- as.vector(which(mydata$y==1))
S3 <- sample(S1, ceiling(length(S1)*0.8), replace=FALSE)
S4 <- sample(S2, ceiling(length(S2)*0.8), replace=FALSE)
TrainInd <- c(S3, S4)
TestInd <- setdiff(1:length(mydata$y), TrainInd)
TrainXY <- mydata[TrainInd, ]
TestXY <- mydata[TestInd, ]
# Fit a bagging LASSO linear regression model, where the parameters
# of M in the following example is set as small values to reduce the
# running time, however the default value is proposed.
Bagging.fit <- Bagging.lasso(x=TrainXY[, -10], y=TrainXY[, 10],
family=c("gaussian"), M=2, predictor.subset=round((9/10)*ncol(x)),
predictor.importance=TRUE, trimmed=FALSE, weighted=TRUE, seed=0123)
Bagging.fit
# Make predictions from a bagging LASSO linear regression model.
pred <- Predict.bagging(Bagging.fit, newx=TestXY[, -10], y=NULL)
pred