Usage
kaps.control(pre.pt = list(), scope = list(),
rho = 0, V = 5, ncl = 1, lower.limit = 0, upper.limit = 100,
shortcut = TRUE, N.perm = 9999, N.boot = 200, alpha = 0.05,
splits = c("logrank", "exact"),
correct = c("Adj.Bf", "Bf", "None"),
p.adjust.methods = c("none", "holm", "hochberg",
"hommel", "bonferroni", "BH", "BY", "fdr"))Arguments
pre.pt
a parameter with the type of list in order to treat pre-specified split candidates. Use the option as list(var = split points), i.e., x = 1:100
scope
a parameter with the type of list in order to treat pre-determined split range. Use the option as list(var = ranges), i.e., x = c(1,100)
rho
a scalar parameter that controls the type of logrank test. See survdiff. V
a numeric parameter that determine the number of folds in the cross-validation subgroup selection.
ncl
a scalar parameter with the type of integer to determine the number of cores used in parallel computing
lower.limit
a scalar parameter with the type of numeric to treat pre-determined overall lower bound. Default is 0.
upper.limit
a scalar parameter with the type of numeric to treat pre-determined overall upper bound. Default is 100.
shortcut
a logical value. If TRUE, we skip the off-diagonal matrix in pairwise-comparison to reduce computational cost. Default is TRUE.
N.perm
a scalar parameter with the type of numeric to give the number of permutation samples used in the kaps with single permutation, and bootstrap and permuting kaps algorithms. The default value is 9999.
N.boot
a scalar parameter with the type of numeric to give the number of Bootstrap samples used in the bootstrap and permuting kaps algorithm. The default value is 200.
alpha
a scalar parameter with the type of numeric to proviede significant in the process of Bootstrap and permuting
splits
a character value to determine the kind of pairwise test. Default is logrank test. At this stage, the option exact is not working.
correct
a character parameter to select the criteria for the multiple comparison in the simple permuting kaps algorithm.
p.adjust.methods
a character parameter to select the criteria for the multiple comparison.