irboost: Iteratively Reweighted Boosting for Robust Analysis

Fit a predictive model with the iteratively reweighted boosting (IRBoost) that minimizes the robust loss functions in the CC-family (concave-convex). The convex optimization is conducted by functional descent boosting algorithm in the R package xgboost. The IRBoost reduces the weight of the observation that leads to a large loss; it also provides weights to help identify outliers. Applications include the robust generalized linear models and extensions, where the mean is related to the predictors by boosting, and robust accelerated failure time models. The package supersedes the R package ccboost. Wang (2021) <arXiv:2101.07718>.

Version: 0.1-1.1
Depends: R (≥ 3.5.0)
Imports: mpath (≥ 0.4-2.21), xgboost
Suggests: R.rsp, DiagrammeR, survival, Hmisc
Published: 2022-02-16
Author: Zhu Wang ORCID iD [aut, cre]
Maintainer: Zhu Wang <zhuwang at>
License: GPL (≥ 3)
NeedsCompilation: no
Materials: README NEWS
CRAN checks: irboost results


Reference manual: irboost.pdf
Vignettes: Unified Robust Boosting


Package source: irboost_0.1-1.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): irboost_0.1-1.1.tgz, r-oldrel (arm64): irboost_0.1-1.1.tgz, r-release (x86_64): irboost_0.1-1.1.tgz, r-oldrel (x86_64): irboost_0.1-1.1.tgz


Please use the canonical form to link to this page.