Tu & Zhou (1999)
In datasets whose data-generating distributions are non-negative with excess zero observations, it can be difficult to find general-purpose statistical tests for comparing sample means while controlling type-I error rates. This R package allows users to perform a modified bootstrap-based t-test that aims to better control type-I error rates in these particular datasets.
To download and use this package, run the following:
install.packages("devtools") # Unless you already have itlibrary(devtools)devtools::install_github("WannabeSmith/rbtt")
To obtain details on how to use the package, run: