predict.Owl: Generate predictions from owl models

In jolars/prague: Generalized Linear Models Regularized with the Sorted L1-Norm

Description

Return predictions from models fit by owl().

Usage

## S3 method for class 'Owl'
predict(object, x, sigma = NULL, type = "link", simplify = TRUE, ...)

## S3 method for class 'OwlGaussian'
predict(
  object,
  x,
  sigma = NULL,
  type = c("link", "response"),
  simplify = TRUE,
  ...
)

## S3 method for class 'OwlBinomial'
predict(
  object,
  x,
  sigma = NULL,
  type = c("link", "response", "class"),
  simplify = TRUE,
  ...
)

## S3 method for class 'OwlPoisson'
predict(
  object,
  x,
  sigma = NULL,
  type = c("link", "response"),
  exact = FALSE,
  simplify = TRUE,
  ...
)

## S3 method for class 'OwlMultinomial'
predict(
  object,
  x,
  sigma = NULL,
  type = c("link", "response", "class"),
  exact = FALSE,
  simplify = TRUE,
  ...
)

Arguments

object	an object of class "owl", typically the result of a call to owl()
x	new data
sigma	penalty parameter for SLOPE models; if NULL, the values used in the original fit will be used
type	type of prediction; "link" returns the linear predictors, "response" returns the result of applying the link function, and "class" returns class predictions.
simplify	if TRUE, base::drop() will be called before returning the coefficients to drop extraneous dimensions
...	ignored and only here for method consistency
exact	if TRUE and the given parameter values differ from those in the original fit, the model will be refit by calling stats::update() on the object with the new parameters. If FALSE, the predicted values will be based on interpolated coefficients from the original penalty path.

Value

Predictions from the model with scale determined by type.

See Also

stats::predict(), stats::predict.glm()

Examples

fit <- with(mtcars, owl(cbind(mpg, hp), vs, family = "binomial"))
predict(fit, with(mtcars, cbind(mpg, hp)), type = "class")

jolars/prague documentation built on March 4, 2020, 7:13 p.m.