Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

syn.survctree

Synthesis of survival time by classification and regression trees (CART)


Description

Generates synthetic event indicator and time to event data using classification and regression trees (without or with bootstrap).

Usage

syn.survctree(y, yevent, x, xp, proper = FALSE, minbucket = 5, ...)

Arguments

y

a vector of length n with original time data.

yevent

a vector of length n with original event indicator data.

x

a matrix (n x p) of original covariates.

xp

a matrix (k x p) of synthesised covariates.

proper

for proper synthesis (proper = TRUE) a CART model is fitted to a bootstrapped sample of the original data.

minbucket

the minimum number of observations in any terminal node. See ctree_control for details.

...

additional parameters passed to ctree.

Details

The procedure for synthesis by a CART model is as follows:

  1. Fit a tree-structured survival model by binary recursive partitioning (the terminal nodes include Kaplan-Meier estimates of the survival time).

  2. For each xp find the terminal node.

  3. Randomly draw a donor from the members of the node and take the observed value of yevent and y from that draw as the synthetic values.

NOTE that when the function is called by setting elements of method in syn() to "survctree", the parameter minbucket can be supplied to syn() as e.g. survctree.minbucket.

Value

A list with the following components:

syn.time

a vector of length k with synthetic time values.

syn.event

a vector of length k with synthetic event indicator values.

fit

the fitted model which is an item of class ctree.object.

See Also


synthpop

Generating Synthetic Versions of Sensitive Microdata for Statistical Disclosure Control

v1.6-0
GPL-2 | GPL-3
Authors
Beata Nowok [aut, cre], Gillian M Raab [aut], Chris Dibben [ctb], Joshua Snoke [ctb], Caspar van Lissa [ctb]
Initial release
2020-09-03

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.