sim.data.MatchCR: Function to simulate matchced survival data in competing risk...
In cribosch/matchsurv: Excess Risk Estimation for Matched Cohort Survival Data

Description Usage Arguments Value Author(s)

Function to simulate matchced survival data in competing risk setting based on cumulative incidence functions

1
2
3

sim.data.MatchCR(nca, ncont = 5, cifs = NULL, gammax = c(0.1, -0.2),
  mean.link = "log", bias = FALSE, cens = FALSE, cens.dist = FALSE,
  age.expo = NULL, print.cifs = FALSE)

`nca`	number of exposed individuals
`ncont`	number of unexposed individuals for each exposed (fixed number; by default ncont=5)
`cifs`	list argument. Starting CIFs for: 1) 1st cause background risk, 2) 2nd cause background risk, 3) 1st cause excess risk, 4) 2nd cause excess risk. If cifs=NULL default quantities are used.
`gammax`	coefficients for the excess risk model. Max 2 values to be specified (the first covariate is by default binomial, the second is log normal), Default values: γ_1=0.1, \; γ_2=-0.2. Note: some value might not work.
`mean.link`	link function for the gee model. The data are simulating accordingly to the link that will be used to estimate the excess cif model. Two values supported "id" or "log".
`bias`	to simulate data with excess risk factor correlated with age
`cens`	TRUE when simulating data with uniform censoring time (it's the same for exposed and unexposed in the same cluster to mimic the registry structure)
`cens.dist`	TRUE when censoring time depends on X1. FALSE if no censoring or independent censoring times
`age.expo`	NULL if age at entry randomly generated from uniform(0,20). If fix age, specify single number (it can be 0, no delay entry)
`print.cifs`	if information about the given cifs is needed

The function returns by default a dataset. if print.cifs=TRUE, the function returns a list withe the simulated dataset and a list of the startingc cumulative incidence functions.

Cristina Boschini

cribosch/matchsurv documentation built on Aug. 15, 2019, 11:55 a.m.