The binomialDP
package makes it easy to perform differentially private inference for a population proportion in R. The package offers a quick solution to find differentially private uniformly most powerful (UMP) tests, private p-values, and private confidence intervals. Also included in the package are the CDF function and the random deviate generator of the Truncated-Uniform-Laplace (Tulap) distribution.
To install the package, run one of the following:
## The latest official version from CRAN install.packages("binomialDP") library(binomialDP) ## The latest development version from Github install.packages("devtools") devtools::install_github("tranntran/binomialDP")
If you use this package for academic purposes, please cite this package along with the paper on differentially private inference for binomial data:
Jordan Awan, Aleksandra Slavkovic and Tran Tran (2020). binomialDP: Differentially Private Inference for Binomial Data. R package version 1.0.0.
Awan, Jordan Alexander, and Aleksandra Slavkovic. “Differentially Private Inference for Binomial Data.” Journal of Privacy and Confidentiality 10, no. 1 (2020).