Source
grf — Athey, Tibshirani & Wager
@misc{grf,
title = {grf},
author = {Athey and Tibshirani and Wager},
howpublished = {\url{https://grf-labs.github.io/grf/}},
note = {Software / documentation}
}Summary by StatsDoge
Non-parametric conditional survival function S(t | X) under right-censoring.
You're looking at a building block — one of the estimators a
workflow uses inside its pipeline.
You reached it from a workflow step; it's used in 1 workflow (listed below).
https://github.com/grf-labs/grf
@
v2.6.1tag
⚠️ Unofficial community write-up of a method from grf-labs/grf (pinned at
v2.6.1). Not affiliated with the grf-labs authors — this summarizes the public documentation for demonstration. All credit & copyright belong to the original authors (Athey, Tibshirani, Wager, et al.).
What it does
Estimates the conditional survival curve S(t | X) (Kaplan–Meier / Nelson–Aalen within adaptive neighborhoods) for right-censored data.
sf <- survival_forest(X, Y, D) # Y time, D event indicator
predict(sf)$predictions # survival curve per unit
Use it for
Baseline survival modelling and as the censoring/nuisance model feeding a causal_survival_forest.
Used in these workflows (1)
-
Causal forest with time-to-event data (survival)
Censoring check → causal survival forest → RMST-scale AIPW ATE → calibration → report.
Discussion (0)
Log in to join the discussion.