Skip to contents

Train an Elastic Net Cox Proportional Hazards Model

Source: R/prognosis.R
en_pro.Rd

Trains a Cox proportional hazards model with Elastic Net regularization using glmnet (with alpha = 0.5).

Usage

en_pro(X, y_surv, tune = FALSE)

Arguments

X

A data frame of features.

y_surv

A survival::Surv object representing the survival outcome.

tune

Logical, whether to perform hyperparameter tuning (currently simplified/ignored for direct cv.glmnet usage which inherently tunes lambda).

Value

A list of class "train" containing the trained glmnet model object, names of features used in training, and model type. The returned object also includes fitted_scores (linear predictor), y_surv, best_lambda, and alpha_val.

Examples

set.seed(42)
n_samples <- 50
n_features <- 10
X_data <- as.data.frame(matrix(rnorm(n_samples * n_features), ncol = n_features))
Y_surv_obj <- survival::Surv(
  time = runif(n_samples, 100, 1000),
  event = sample(0:1, n_samples, replace = TRUE)
)

# Train the model
en_model <- en_pro(X_data, Y_surv_obj)
print(en_model$finalModel)
#> 
#> Call:  glmnet::glmnet(x = X_matrix, y = y_surv, family = "cox", alpha = alpha_val,      lambda = best_lambda) 
#> 
#>   Df %Dev Lambda
#> 1  2 4.27 0.2082