sis.surv: Variable Screening before Selection for Survival Response
In hanfu-bios/varsel: Model-Free Knockoffs Filter for Controlled Variable Selection

Description Usage Arguments Details Value References

When the response is a survival object, this function screens the variables before computing their statistics to reduce computing time, especially when p is large.

1	sis(Xaug, y, N = nrow(Xaug), knot = 10)

`Xaug`	augmented design matrix combining original predictors and knockoff variables
`y`	response vector, or a survival object with two columns
`N`	number of variables left after screening (default: sample size)
`knot`	parameter in screening process

Sure Independence Screening (SIS) method is used in this function. The screening process would accelerate computing but not necessarily improve power as true signals might be removed by this process. Our experiment indicates that the power would remain almost the same.

a binary 2p by 1 vector for screening results. 1 means selected by the screening and 0 otherwise.

Jianqing Fan and Jinchi Lv (2008) Sure Independence Screening for Ultrahigh Dimensional Feature Space (with discussion). Journal of Royal Statistical Society B, 70, 849-911.

hanfu-bios/varsel documentation built on May 27, 2019, 4:50 a.m.