mvs (version 2.0.0)

predict.MVS: Make predictions from an "MVS" object.

Description

Make predictions from a "MVS" object.

Usage

# S3 method for MVS
predict(object, newx, predtype = "response", cvlambda = "lambda.min", ...)

Value

A matrix of predictions.

Arguments

object

An object of class "MVS".

newx

Matrix of new values for x at which predictions are to be made. Must be a matrix.

predtype

The type of prediction returned by the meta-learner. Supported are types "response", "class" and "link".

cvlambda

Values of the penalty parameters at which predictions are to be made. Defaults to the values giving minimum cross-validation error.

...

Further arguments to be passed to predict.cv.glmnet.

Author

Wouter van Loon <[e-mail]>

Examples

# \donttest{ 
set.seed(012)
n <- 1000
X <- matrix(rnorm(8500), nrow=n, ncol=85)
top_level <- c(rep(1,45), rep(2,20), rep(3,20))
bottom_level <- c(rep(1:3, each=15), rep(4:5, each=10), rep(6:9, each=5))
views <- cbind(bottom_level, top_level)
beta <- c(rep(10, 55), rep(0, 30)) * ((rbinom(85, 1, 0.5)*2)-1)
eta <- X %*% beta
p <- 1 /(1 + exp(-eta))
y <- rbinom(n, 1, p)

fit <- MVS(x=X, y=y, views=views, type="StaPLR", levels=3, alphas=c(0,1,1), nnc=c(0,1,1))
coefficients <- coef(fit)

new_X <- matrix(rnorm(2*85), nrow=2)
predict(fit, new_X)# }