survscreen.marg: Wrapper function for marginal Cox regression screening...

In tedwestling/survSuperLearner: Super Learning for Conditional Survival Functions with Right-Censored Data

Wrapper function for marginal Cox regression screening algorithm

Description

This screening algorithm uses marginal coxph regressions to select covariates that have significant marginal relationships with the event.

Usage

survscreen.marg(time, event, X, obsWeights, minscreen = 2, min.p = 0.1, ...)

Arguments

time	Observed follow-up time; i.e. minimum of the event and censoring times.
event	Observed event indicator; i.e, whether the follow-up time corresponds to an event or censoring.
X	Training covariate data.frame.
obsWeights	Observation weights.
minscreen	Minimum number of covariates to return. Defaults to 2.
min.p	Threshold p-value used to decide if a covariate is included. Defaults to 0.1
...	Additional ignored arguments.
alpha	Penalty exponent for glmnet. Defaults to 1 (lasso penalty).

Details

A univariate Cox regression is run for each covariate; those with p-values less than min.p are included.

Value

Logical vector of the same length as the number of columns of X indicating which variables were included.

tedwestling/survSuperLearner documentation built on Dec. 12, 2024, 4:16 p.m.