surv.bart.dss: Decoupling Shrinkage and Selection (DSS) variable selection...

In timebart: timebart: Survival Analysis with BART

Description

Here we implement the Decoupling Shrinkage and Selection (DSS) method for variable selection in nonparametric survival analysis with BART. The method is computationally friendly.

Usage

    surv.bart.dss(post,
                  mc.cores=2L,   ## used by mc.surv.bart.dss only
                  nice=19L       ## used by mc.surv.bart.dss only
                  )
    mc.surv.bart.dss(post,
                     mc.cores=2L,## used by mc.surv.bart.dss only
                     nice=19L    ## used by mc.surv.bart.dss only
                     )

Arguments

post	The posterior from a previous call to surv.bart or mc.surv.bart.
mc.cores	Number of cores to employ in parallel: used by mc.surv.bart.dss only.
nice	Set the job priority: used by mc.surv.bart.dss only. The default priority is 19: priorities go from 0 (highest) to 19 (lowest).

Value

Returns a list of selected variable column numbers and names.

Author(s)

Rodney Sparapani: [e-mail]

References

A General Approach to Variable Selection for Bayesian Nonparametric Models JSM, August 10, 2015. http://rob-mcculloch.org/nonpar-varsel_2015-8-10.pdf

See Also

surv.bart

Examples

## Not run: 
require(timebart)

## End(Not run)

timebart documentation built on May 2, 2019, 4:43 p.m.