predict.randomGLM {randomGLM}

Prediction from a random generalized linear model predictor

Description

Implements a predict method on a previously-constructed random generalized linear model predictor and new data.

Usage

## S3 method for class 'randomGLM'
predict(object, newdata, type=c("response", "class"), 
                 thresholdClassProb = object$details$thresholdClassProb, ...)

Arguments

object

a randomGLM object such as one returned by randomGLM.

newdata

specification of test data for which to calculate the prediction.

type

type of prediction required. Type "response" gives the fitted probabilities for classification, the fitted values for regression. Type "class" applies only to classification, and produces the predicted class labels.

thresholdClassProb

the threshold of predictive probabilities to arrive at classification. Takes values between 0 and 1. Only used for binary outcomes.

...

other arguments that may be passed to and from methods. Currently unused.

Details

The function calculates prediction on new test data. It only works if object contains the regression models that were used to construct the predictor (see argument keepModels of the function randomGLM).

If the predictor was trained on a multi-class response, the prediction is applied to each of the representing binary variables (see randomGLM for details).

Value

For continuous prediction, the predicted values. For classification of binary response, predicted class when type="class"; or a two-column matrix giving the class probabilities if type="response".

If the predictor was trained on a multi-class response, the returned value is a matrix of "cbind"-ed results for the representing individual binary variables (see randomGLM for details).

Author(s)

Lin Song, Steve Horvath and Peter Langfelder.

References

Lin Song, Peter Langfelder, Steve Horvath: Random generalized linear model: a highly accurate and interpretable ensemble predictor. BMC Bioinformatics (2013)

Examples

## binary outcome prediction
# data generation
data(iris)
# Restrict data to first 100 observations
iris=iris[1:100,]
# Turn Species into a factor
iris$Species = as.factor(as.character(iris$Species))
# Select a training and a test subset of the 100 observations
set.seed(1)
indx = sample(100, 67, replace=FALSE)
xyTrain = iris[indx,]
xyTest = iris[-indx,]
xTrain = xyTrain[, -5]
yTrain = xyTrain[, 5]

xTest = xyTest[, -5]
yTest = xyTest[, 5]

# predict with a small number of bags 
# - normally nBags should be at least 100.
RGLM = randomGLM(
   xTrain, yTrain, 
   nCandidateCovariates=ncol(xTrain), 
   nBags=30, 
   keepModels = TRUE, nThreads = 1)

predicted = predict(RGLM, newdata = xTest, type="class")
table(predicted, yTest)

## continuous outcome prediction

x=matrix(rnorm(100*20),100,20)
y=rnorm(100)

xTrain = x[1:50,]
yTrain = y[1:50]
xTest = x[51:100,]
yTest = y[51:100]

RGLM = randomGLM(
   xTrain, yTrain, 
   classify=FALSE, 
   nCandidateCovariates=ncol(xTrain), 
   nBags=10, 
   keepModels = TRUE, nThreads = 1)

predicted = predict(RGLM, newdata = xTest)