• Updated URLs in DESCRIPTION file;
  • Properly documented package overview in flipr-package.R file;
  • Updated GitHub Action workflows;
  • Removed suggested dependency to the entire tidyverse;
  • Updated maintainer email.
  • Removed akima from package dependencies.
  • Updated GHA workflows and roxygen2 version.
  • Updated all URLs to reflect change of ownership.
  • Remove any occurrence of base-R pipe operator in favor of magrittr pipe operator for now for backward compatibility with older versions of R.
  • Improved way of handling statistics that rely on inter-point distances.
  • Method $get_value() gains optional arguments keep_null_distribution and keep_permutations in case the user would like to keep track of either the permutation null distribution or the sampled permutations that have been used to produce it.
  • Method $get_value() now handles extra parameters to be passed to the test statistics.

Breaking changes

The API has been changed:

  • An R6 class is now used to define the plausibility function, with methods that implement point estimation, confidence intervals and evaluation on grid for later plotting.
  • The plotting capability is still outside the class but might move into the class methods as well in the future.

Minor changes

  • Test statistics based on Inter-point distances have been added and implemented in C++ through the Rcpp package.
  • Test coverage is set and achieves 20% coverage for now (#3, @C-Juliette).
  • @C-Juliette has been added as contributor to the package.
  • The p-value function has been consistently renamed plausibility function to better reflects what it stands for and avoid confusion with other definitions of the p-value function in the area of functional data analysis.
  • Fix warning when installing the package on Oracle Solaris 10, x86, 32 bit, R-release due to pandoc version not available preventing the pvalue-function vignette from knitting properly.

New features

  • Compute p-value surfaces when inferring multiple parameters at once via two_sample_pf();
  • Compute a point estimation for a single parameter of interest via two_sample_pe();
  • Compute a confidence interval for a single parameter of interest via two_sample_ci();
  • Automatically draw the p-value function for a single parameter of interest via two_sample_viz().

Full list of changes

  • Change argument names test and combining_function to type and combine_with respectively. Add argument alternative to specify which kind of test is performed.
  • Add a function to compute p-value functions for sets of null hypotheses.
  • Draft of article illustrating the computation of p-value functions with flipr.
  • Add t, mean and Fisher test statistics.
  • Correct two-tail p-value computation.
  • Better API for pvalue function.
  • Added a function for two sample confidence interval for a single parameter.
  • Added an article about permutation inference overview.
  • Added dependency to the ggplot2 package for plotting features.
  • Added dependency to the cli package for displaying important information to the user in an elegant fashion.
  • Added function for point estimation of a single parameter.
  • Added function for confidence interval of a single parameter.
  • Added function for automatic plotting of the p-value function of a single parameter.
  • Added dependency to the withr package for properly handle local random seed generator settings.
  • Added dependency to the viridisLite package for using colors compatible with most common types of color blindness.
  • Fix warning when installing the package on Oracle Solaris 10, x86, 32 bit, R-release due to pandoc version not available.
  • Switch to GPL-3 license because purrr dependency is GPL-3 licensed.
  • Initial release.
  • Added a NEWS.md file to track changes to the package.