iid.prodlim: Extract i.i.d. decomposition from a prodlim model

In bozenne/BuyseTest: Generalized Pairwise Comparisons

Extract i.i.d. decomposition from a prodlim model

Description

Compute the influence function for each observation used to estimate the model

Usage

## S3 method for class 'prodlim'
iid(x, add0 = FALSE, ...)

Arguments

x	A prodlim object.
add0	[logical] add the 0 to vector of relevant times.
...	not used. For compatibility with the generic method.

Details

This function is a simplified version of the iidCox function of the riskRegression package. Formula for the influence function can be found in (Ozenne et al., 2017).

Value

A list containing:

• IFbeta: Influence function for the regression coefficient.

• IFhazard: Time differential of the influence function of the hazard.

• IFcumhazard: Influence function of the cumulative hazard.

• time: Times at which the influence function has been evaluated.

• etime.max: Last observation time (i.e. jump or censoring) in each strata.

• label.strata: Strata to which each observation belong.

• X: Design matrix.

• table: Hazard at each time for each strata.

Author(s)

Brice Ozenne

References

Brice Ozenne, Anne Lyngholm Sorensen, Thomas Scheike, Christian Torp-Pedersen and Thomas Alexander Gerds. riskRegression: Predicting the Risk of an Event using Cox Regression Models. The R Journal (2017) 9:2, pages 440-460.

Examples

library(data.table)
library(prodlim)

set.seed(10)
dt <- simBuyseTest(10)
setkeyv(dt, "treatment")

e.KM <- prodlim(Hist(eventtime,status)~treatment, data = dt)
lava::iid(e.KM)

bozenne/BuyseTest documentation built on Feb. 16, 2024, 5:35 a.m.