SL.glm: Wrapper for glm

In SuperLearner: Super Learner Prediction

Wrapper for glm

Description

Wrapper for generalized linear models via glm().

Note that for outcomes bounded by [0, 1] the binomial family can be used in addition to gaussian.

Usage

SL.glm(Y, X, newX, family, obsWeights, model = TRUE, ...)

Arguments

Y	Outcome variable
X	Training dataframe
newX	Test dataframe
family	Gaussian or binomial
obsWeights	Observation-level weights
model	Whether to save model.matrix of data in fit object. Set to FALSE to save memory.
...	Any remaining arguments, not used.

References

Fox, J. (2015). Applied regression analysis and generalized linear models. Sage Publications.

See Also

predict.SL.glm glm predict.glm SL.speedglm

Examples

data(Boston, package = "MASS")
Y = Boston$medv
# Remove outcome from covariate dataframe.
X = Boston[, -14]

set.seed(1)

sl = SuperLearner(Y, X, family = gaussian(),
                  SL.library = c("SL.mean", "SL.glm"))

print(sl)

SuperLearner documentation built on May 29, 2024, 5:25 a.m.