predict.sgp.cv {SGPR}

Predictions based on a SGP models

Description

A function that extracts information from a cross-validated SGP object and performs predictions.

Usage

## S3 method for class 'sgp.cv'
predict(
  object,
  X,
  lambda = object$lambda.min,
  index = object$min,
  extract = c("link", "response", "class", "coefficients", "vars", "groups", "nvars",
    "ngroups", "norm"),
  ...
)

Arguments

object

A object that was generated with sgp.cv.

X

The design matrix for making predictions.

lambda

The value of lambda at which predictions should be made.

index

The index that indicates the lambda at which predictions should be made (alternative to specifying 'lambda').

extract

A string indicating the type of information to return.

...

Other parameters of underlying basic functions.

Value

Different objects depending on the sting indicated by 'extract'.

Examples

n <- 100
p <- 12
nr <- 4
g <- paste0("Group ",ceiling(1:p / nr))
X <- matrix(rnorm(n * p), n, p)
b <- c(-3:3)
y_lin <- X[, 1:length(b)] %*% b + 5 * rnorm(n)
y_log <- rbinom(n, 1, exp(y_lin) / (1 + exp(y_lin)))

lin_fit <- sgp.cv(X, y_lin, g, type = "linear")
predict(lin_fit, X = X, extract = "link")

log_fit <- sgp.cv(X, y_log, g, type = "logit")
predict(log_fit, X = X, extract = "class")