Description
Blyth-Still-Casella Confidence Interval.
Description
Provides a fast calculation of the Blyth-Still-Casella confidence interval. The implementation follows the 'StatXact' 9 manual (Cytel 2010) and "Refining Binomial Confidence Intervals" by George Casella (1986) <doi:10.2307/3314658>.
README.md
rbscCI
This package is an R wrapper of the C++ implementaton of the Blyth-Still-Casella confidence interval https://github.com/keithw/biostat.
It requires C++11 and the boost C++ library: https://www.boost.org/. The original C++ code is wrapped with an R interface using Rcpp and the boost library depedency is handled by the BH R package.
Installation
You can install the released version of rbscCI from CRAN with:
install.packages("rbscCI")
Example
This is a basic example to find the 95% confidence interval of the probability of success when there are 25 success out of 100 coin flips:
library(rbscCI)
bscCI(100, 25, 0.95, digits = 3)
#> [1] 0.171 0.341