This package offers a user friendly function ‘crrcbcv’ to compute bias-corrected variances for competing risks regression models using proportional subdistribution hazards with small-sample clustered data. Four types of bias correction are included: the MD-type bias correction by Mancl and DeRouen (2001), the KC-type bias correction by Kauermann and Carroll (2001), the FG-type bias correction by Fay and Graubard (2001), and the MBN-type bias correction by Morel, Bokossa, and Neerchal (2003).
You can install the released version of crrcbcv from CRAN with:
install.packages("crrcbcv")
This is a basic example which shows you how to solve a common problem:
library(crrcbcv)
#> Loading required package: crrSC
#> Loading required package: survival
#> Loading required package: abind
#> Loading required package: pracma
data(cls)
= crrc(ftime=cls$T_obs, fstatus=cls$eps, cov1=cls[,c('X_1','X_2')], cluster=cls$I)
mod.est crrcbcv(beta=mod.est$coef, ftime=cls$T_obs, fstatus=cls$eps, cov1=cls[,c('X_1','X_2')],
cluster=cls$I, var.type=c('MD','KC','FG','MBN'))
#> $MD
#> [,1] [,2]
#> [1,] 0.18712217 0.013304534
#> [2,] 0.01330453 0.007767604
#>
#> $KC
#> [,1] [,2]
#> [1,] 0.14807541 0.155472768
#> [2,] 0.01351378 0.006116421
#>
#> $FG
#> [,1] [,2]
#> [1,] 0.145317028 0.005252785
#> [2,] 0.005252785 0.005509824
#>
#> $MBN
#> [,1] [,2]
#> [1,] 0.133946021 0.004038105
#> [2,] 0.004038105 0.008991263
#>
#> attr(,"class")
#> [1] "crrcbcv"