fittrees: Fit Trees for 5-STAR

In rmarceauwest/fiveSTAR: Run 5-STAR Algorithm

Fit Trees for 5-STAR

Description

Fits preliminary (3A) and pooled final (3B) trees to determine homogenous risk strata for use with 5-STAR algorithm

Usage

fittrees(yy, X, family = "cox", verbose = 0,
  tree.hyper = tree_control())

Arguments

yy	Trait/response (either a binary or continuous covariate or a Surv() object summarizing follow-up time for right-censored data and status indicator where 1=dead, 0=censored)
X	Data frame of all possible stratification covariates
family	Trait family, current options: "cox", "binomial", or "gaussian"
verbose	Numeric variable indicating amount of information to print to the terminal (0 = nothing, 1 = notes only, 2+ = notes and intermediate output)
tree.hyper	List of control variables for tree fitting (see tree_control for details), many of which will be passed into ctree_control.

Value

• strataids: vector of subject-level strata membership (1 = highest risk, 2 = 2nd highest risk, etc.)

• stratadefn: definition of final formed strata, in terms of covariates

• finaltree: ctree object; final, pruned tree built by ctree, in terms of preliminary strata membership

• prelimstrataids: vector of subject-level strata membership for preliminary, unpruned tree

• prelimstratadefn: definition of final formed strata, in terms of covariates for preliminary, unpruned tree

• prelimtree: ctree object; preliminary, unpruned tree built by ctree, in terms of covariates

rmarceauwest/fiveSTAR documentation built on June 30, 2023, 7:38 a.m.