This software is longer maintained and has been removed from CRAN. I recommend using the Generalized Random Forest package for R instead: https://github.com/swager/grf
R package for generating prediction and prediction variances from random forests. Read the vignettes for an introduction to the package (vignette('RFinfer-examples')
) and the theory behind the infinitesimal jackknife applied to random forest predictions (vignette('IJ_vig')
).
Options for random forest include using a subsample or bootstrap sample for each tree and using either conditional inference tree or traditional CART used as base learners.
Install from CRAN: install.packages('RFinfer')
Install latest version: devtools::install_github('cole-brokamp/RFinfer')
This package is based on simulation experiments in: Cole Brokamp, MB Rao, Patrick Ryan, Roman Jandarov. A comparison of resampling and recursive partitioning methods in random forest for estimating the asymptotic variance using the infinitesimal jackknife. Stat. 6(1). 360-372. 2017. Download.