Sparse-group-lasso Cox backend

Wrappers for Cox models fitted with SGL.

Usage

fit.SGL.cox(response, x, cplx, index, ...)
complexity.cvSGL.cox(response, x, full.data, index, ...)
# S3 method for class 'SGL_cox'
predictProb(object, response, x, times, complexity = NULL, ...)
# S3 method for class 'SGL_cox'
PLL(object, newdata, newtime, newstatus, complexity = NULL, ...)

Arguments

response

survival response as a Surv object or a two-column time/status matrix.

x

covariate matrix.

cplx

selected lambda value.

index

group membership vector for the columns of x.

full.data

full data frame, accepted for the peperr complexity-function contract.

object

a fitted SGL_cox object.

times

evaluation times for survival probabilities.

complexity

selected lambda value.

newdata

new covariate matrix.

newtime

vector of follow-up times.

newstatus

vector of event indicators.

...

additional arguments passed to SGL or cvSGL.

Value

Fitted SGL_cox objects, selected lambda values, survival-probability matrices, and numeric predictive partial log-likelihood values, respectively.

Details

The SGL predictor returns relative risks. The peperr wrapper reconstructs survival probabilities from those risks with a Breslow baseline estimated on the stored training data.

See also

peperr, SGL, cvSGL