pmdhvp: Interaction Term Ranking with PMD criteria

In yifansha/highinthunt: High Order Interaction Variable Importance Through Random Forest

Description

This function ranks the interaction terms in a rfsrc object according to pairwise minimal depth (PMD) matrix.

Usage

pmdhvp(obj, inter, verbose = T)

Arguments

obj	An object of class (rfsrc, grow).
inter	A list of vectors that store combinations in interaction terms.
verbose	Set to TRUE for verbose output.

Value

pmdvp	PMD variable importance for each interaction terms in inter.
pmd	PMD matrix calculated from input obj.

Author(s)

Yifan Sha and Min Lu

References

Ishwaran H. (2007). Variable importance in binary regression trees and forests, Electronic J. Statist., 1:519-537.

Ishwaran H., Kogalur U.B., Gorodeski E.Z, Minn A.J. and Lauer M.S. (2010). High-dimensional variable selection for survival data. J. Amer. Statist. Assoc., 105:205-217.

Ishwaran H., Kogalur U.B., Chen X. and Minn A.J. (2011). Random survival forests for high-dimensional data. Statist. Anal. Data Mining, 4:115-132.

Examples

data(express)
o.0 <- rfsrc(y~., data = express[,1:11])
## computing combinations up to 5 way interaction terms
cmbn <- lapply(2:5,function(i){t(combn(1:length(o.0$xvar.names),i))})
inter <- unlist(lapply(1:length(cmbn), function(i){
     lapply(1:nrow(cmbn[[i]]),function(j){
     matrix(cmbn[[i]][j,],1,length(cmbn[[i]][j,])) })
                      }), recursive = FALSE )

o <- pmdhvp(obj = o.0, inter, verbose = TRUE)
o$pmdvp
## rank(o$pmdvp)[which(rownames(o$pmdvp) == "x1578_x1430_x692_x1223")]

yifansha/highinthunt documentation built on July 2, 2020, 6:29 p.m.