predict.coxkl_enet: Predict Linear Predictors from a coxkl_enet Object

In survkl: Estimate Survival Data with Data Integration

Predict Linear Predictors from a coxkl_enet Object

Description

Computes linear predictors for new data using a fitted coxkl_enet model. If lambda is supplied, predictions are returned for those lambda values; otherwise predictions are returned for all fitted lambdas. When a requested lambda lies between fitted values, coefficients are linearly interpolated.

Usage

## S3 method for class 'coxkl_enet'
predict(object, newz, lambda = NULL, ...)

Arguments

object	A fitted model object of class "coxkl_enet".
newz	A numeric matrix or data frame of new covariates (same columns as in training data).
lambda	Optional numeric value(s) specifying the regularization parameter(s) for which to predict. If NULL, predictions for all fitted lambda values are returned.
...	Additional arguments.

Details

The linear predictors are computed as as.matrix(newz) %*% beta.

Value

A numeric matrix of linear predictors. Each column corresponds to one lambda, sorted in descending order.

See Also

coef.coxkl_enet

survkl documentation built on April 22, 2026, 1:08 a.m.