A package for doing permutations and bootstraping easily in R. This package provide a means for hypothesis testing when the data does not follow the assumtions of parametric test. A permutation test is done by doing resampling of the observed data without replacement, thereby creating the null-distribution for the association between two variables. Bootstraping is done by doing resampling of the observed data woth replacement, thereby creating the sampling distribution of a desirered estimate, e.g the difference in mean. From the Sampling distribution, the standart error and confidence intervals are calculated.
The permutation part of the package returns a plot of the distribution under the null-hypotesis and a p-value.
The bootstrap will return a plot of the sampling distribution of a given estimate, it's standard error and confidence intervals.
Lotte hindhede Cornelia Buus
Project Manager (pm) Lotte Hindhede
Documentation Executive (de) Cornelia Buus
Quality Executive (qe) Cornelia Buus