Provides an interpretable identification of subgroups with heterogeneous causal effect. The heterogeneous subgroups are discovered through ensemble learning of causal rules. Causal rules are highly interpretable if-then statement that recursively partition the features space into heterogeneous subgroups. A small number of significant causal rules are selected through Stability Selection to control for family-wise error rate in the finite sample setting. It proposes various estimation methods for the conditional causal effects for each discovered causal rule. It is highly flexible and multiple causal estimands and imputation methods are implemented. Lee, K., Bargagli-Stoffi, F. J., & Dominici, F. (2020). Causal rule ensemble: Interpretable inference of heterogeneous treatment effects. arXiv preprint <arXiv:2009.09036>.
|Depends:||R (≥ 3.5.0)|
|Imports:||MASS, stats, logger, gbm, randomForest, methods, xgboost, RRF, data.table, xtable, glmnet, bartCause, stabs, stringr, SuperLearner, magrittr, ggplot2, inTrees|
|Suggests:||baggr, grf, BART, gnm, covr, knitr, rmarkdown, testthat (≥ 3.0.0)|
|Author:||Naeem Khoshnevis [aut, cre] (FASRC), Daniela Maria Garcia [aut], Riccardo Cadei [aut], Kwonsang Lee [aut], Falco Joannes Bargagli Stoffi [aut]|
|Maintainer:||Naeem Khoshnevis <nkhoshnevis at g.harvard.edu>|
|CRAN checks:||CRE results|
Testing the CRE package
|Windows binaries:||r-devel: CRE_0.2.3.zip, r-release: CRE_0.2.3.zip, r-oldrel: CRE_0.2.3.zip|
|macOS binaries:||r-release (arm64): CRE_0.2.3.tgz, r-oldrel (arm64): CRE_0.2.3.tgz, r-release (x86_64): CRE_0.2.3.tgz, r-oldrel (x86_64): CRE_0.2.3.tgz|
|Old sources:||CRE archive|
Please use the canonical form https://CRAN.R-project.org/package=CRE to link to this page.