The Prognostic Regression Offsets with Propagation of ERrors (for Treatment Effect Estimation) package facilitates direct adjustment for experiments and observational studies that is compatible with a range of study designs and covariance adjustment strategies. It uses explicit specification of clusters, blocks and treatment allocations to furnish probability of assignment-based weights targeting any of several average treatment effect parameters, and for standard error calculations reflecting these design parameters. For covariance adjustment of its Hajek and (one-way) fixed effects estimates, it enables offsetting the outcome against predictions from a dedicated covariance model, with standard error calculations propagating error as appropriate from the covariance model.
Version: | 1.0.0 |
Depends: | R (≥ 4.1.0) |
Imports: | stats, methods, sandwich |
Suggests: | knitr, rmarkdown, testthat (≥ 3.0.0), multcomp, MASS |
Enhances: | robustbase |
Published: | 2025-08-21 |
DOI: | 10.32614/CRAN.package.propertee |
Author: | Josh Errickson [cre, aut], Josh Wasserman [aut], Mark Fredrickson [ctb], Adam Sales [ctb], Xinhe Wang [ctb], Ben Hansen [aut] |
Maintainer: | Josh Errickson <jerrick at umich.edu> |
BugReports: | https://github.com/benbhansen-stats/propertee/issues |
License: | MIT + file LICENSE |
URL: | https://github.com/benbhansen-stats/propertee |
NeedsCompilation: | no |
Materials: | NEWS |
CRAN checks: | propertee results |
Reference manual: | propertee.html , propertee.pdf |
Vignettes: |
Regression Discontinuity Design (source, R code) Introduction to propertee (source, R code) Non-binary Treatment Specification (source, R code) |
Package source: | propertee_1.0.0.tar.gz |
Windows binaries: | r-devel: not available, r-release: not available, r-oldrel: not available |
macOS binaries: | r-release (arm64): not available, r-oldrel (arm64): not available, r-release (x86_64): not available, r-oldrel (x86_64): not available |
Please use the canonical form https://CRAN.R-project.org/package=propertee to link to this page.