README.md

In jbryer/PSAboot: Bootstrapping for Propensity Score Analysis

Bootstrapping Propensity Score Analysis

Package website: https://jbryer.github.io/PSAboot/ Poster: https://github.com/jbryer/PSAboot/blob/master/Poster/PSAboot_Poster.pdf

As the popularity of propensity score methods for estimating causal effects in observational studies increase, the choices researchers have for which methods to use has also increased. Estimated treatment effects may be sensitive to choice of method. One approach to test the sensitivity of method choice is to test the null hypothesis more than once using more than one method (Rosenbaum, 2012). With the wide availability of high power computers resampling methods such as bootstrapping (Efron, 1979) have become popular for providing more estimates of the sampling distribution. This paper introduces the PSAboot R package that provides functions for bootstrapping propensity score methods. It deviates from traditional bootstrapping methods by allowing for different sampling specifications for treatment and control groups, mainly to ensure the ratio of treatment-to-control observations are consistent. This approach can also be used in situations where there is imbalance between the number of treatment and control observations by allowing for up and/or down sampling. Lastly, by estimating balance statistics and treatment effects for each bootstrap sample we can compare the distributions across multiple propensity score methods to examine the relative performance of these methods.

Installation

You can download from CRAN using:

install.packages('PSAboot')

Or the latest development version using the remotes package:

remotes::install_github('jbryer/PSAboot')

Code of Conduct

Please note that the PSAboot project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

jbryer/PSAboot documentation built on Oct. 29, 2023, 10 a.m.