wt.itr: Iteratively Grows Weighted Random Forests

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

Description

This function iteratively grows Random Forests using pairwise minimal depth (PMD) weights.

Usage

wt.itr(formula, data,
       pmd.initial,
       obj.initial = NULL,
       wt = function(btpmd,digpmd){log(1/btpmd/digpmd)},
       iteration = 3)

Arguments

formula	A symbolic description of the model to be fit.
data	Data frame containing the y-outcome and x-variables.
pmd.initial	Inital PMD matrix. Note that variables should be arranged as the same order as data.
obj.initial	Inital object of class (rfsrc, grow). New object will be created using the data and formula if it is set to null.
wt	A function calculating variable weights using PMD matrix where btpmd[i] is the average of ith row in the PMD matrix where only off diagonal elements that have smaller values are used and digpmd[i] is the ith diagonal element in the PMD matrix.
iteration	Number of iterations.

Value

A list of each iteration's output:

pmd	PMD matrix after jth iteration (first p columns in the dataframe. When there is only one iteration, this equals to pmd.initial.
digpmd	Diagonal elements of the PMD matrix used in the jth iteration.
btpmd	The average of ith row in the PMD matrix used in the jth iteration, where only off diagonal elements that have smaller values are used.
wt	Variable weights used in the jth iteration.

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])
pmd.0 <- find.interaction(o.0, method="maxsubtree", sorted = FALSE, verbose = FALSE)

o <- wt.itr(y~., data = express[,1:11],
            pmd.initial = pmd.0,
            wt = function(btpmd,digpmd){log(1/btpmd/digpmd)},
            iteration = 2)
o[[2]]

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