Cox Proportional Hazards

Calls survival::coxph() during training. During prediction, the linear predictor is estimated using survival::predict.coxph() while the survival distribution using the Breslow estimator using survival::survfit.coxph()

Dictionary

This Learner can be instantiated via the dictionary mlr_learners or with the associated sugar function lrn():

mlr_learners$get("surv.coxph")
lrn("surv.coxph")

Meta Information

• Task type: “surv”

• Predict Types: “crank”, “distr”, “lp”

• Feature Types: “logical”, “integer”, “numeric”, “factor”

• Required Packages: mlr3, mlr3survival, survival

Parameters

Id	Type	Default	Levels	Range
ties	character	efron	efron, breslow, exact	-
singular.ok	logical	TRUE	TRUE, FALSE	-
type	character	efron	efron, aalen, kalbfleisch-prentice	-
stype	integer	2		\([1, 2]\)

References

Cox DR (1972). “Regression Models and Life-Tables.” Journal of the Royal Statistical Society: Series B (Methodological), 34(2), 187–202. doi:10.1111/j.2517-6161.1972.tb00899.x .

See also

Other survival learners: mlr_learners_surv.kaplan

Super classes

mlr3::Learner -> mlr3survival::LearnerSurv -> LearnerSurvCoxPH

Methods

Public methods

• LearnerSurvCoxPH$new()

• LearnerSurvCoxPH$clone()

Inherited methods

• mlr3::Learner$base_learner()
• mlr3::Learner$configure()
• mlr3::Learner$encapsulate()
• mlr3::Learner$format()
• mlr3::Learner$help()
• mlr3::Learner$predict()
• mlr3::Learner$predict_newdata()
• mlr3::Learner$print()
• mlr3::Learner$reset()
• mlr3::Learner$selected_features()
• mlr3::Learner$train()

---

Method new()

Creates a new instance of this R6 class.

Usage

LearnerSurvCoxPH$new()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LearnerSurvCoxPH$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# Define the Learner
learner = lrn("surv.coxph")
learner
#> 
#> ── <LearnerSurvCoxPH> (surv.coxph): Cox Proportional Hazards ───────────────────
#> • Model: -
#> • Parameters: list()
#> • Packages: mlr3, mlr3survival, and survival
#> • Predict Types: [crank], distr, and lp
#> • Feature Types: logical, integer, numeric, and factor
#> • Encapsulation: none (fallback: -)
#> • Properties: weights
#> • Other settings: use_weights = 'use'

# Define a Task
task = tsk("lung")

# Stratification based on event
task$set_col_roles(cols = "status", add_to = "stratum")

# Create train and test set
part = partition(task)

# Train the learner on the train set
learner$train(task, row_ids = part$train)
learner$native_model
#> Call:
#> survival::coxph(formula = task$formula(), data = task$data(), 
#>     model = TRUE)
#> 
#>                 coef  exp(coef)   se(coef)      z       p
#> age        0.0112226  1.0112858  0.0142049  0.790 0.42950
#> meal.cal  -0.0001446  0.9998554  0.0004184 -0.346 0.72970
#> pat.karno -0.0167315  0.9834077  0.0098372 -1.701 0.08897
#> ph.ecog    0.7279428  2.0708161  0.2806164  2.594 0.00948
#> ph.karno   0.0274259  1.0278055  0.0136580  2.008 0.04464
#> sexm       0.3791073  1.4609798  0.2420168  1.566 0.11724
#> wt.loss   -0.0264181  0.9739278  0.0114005 -2.317 0.02049
#> 
#> Likelihood ratio test=17.94  on 7 df, p=0.01225
#> n= 112, number of events= 81 

# Make predictions for the test set
predictions = learner$predict(task, row_ids = part$test)
predictions
#> 
#> ── <PredictionSurv> for 56 observations: ───────────────────────────────────────
#>  row_ids time status      crank         lp     distr
#>        4  310   TRUE  0.8737011  0.8737011 <list[1]>
#>        6  218   TRUE -0.2003632 -0.2003632 <list[1]>
#>        7  166   TRUE  0.3892358  0.3892358 <list[1]>
#>      ---  ---    ---        ---        ---       ---
#>      163  197  FALSE  0.3359723  0.3359723 <list[1]>
#>      164  203  FALSE  0.1232581  0.1232581 <list[1]>
#>      167  174  FALSE  0.5195508  0.5195508 <list[1]>

# Score the predictions
predictions$score()
#> surv.cindex 
#>   0.6748682