Stata package to perform randomization inference on (almost) any stata command
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.gitattributes
.gitignore
README.md
example_classes.dta
example_students.dta
ritest.ado
ritest.bib
ritest.pdf
ritest.pkg
ritest.sthlp
stata.toc

README.md

ritest

Stata package to perform randomization inference on any stata command.

The first version was published alongside an article in the Stata Journal. I provide provide regular updates through GitHub.

Installation

To obtain the latest version through github, from the main window in Stata, run:

net describe ritest, from(https://raw.githubusercontent.com/simonheb/ritest/master/)

If the download from within Stata fails (e.g. because you are behind a firewall),you can always download the files directly:

Citation

Heß, Simon, "Randomization inference with Stata: A guide and software" Stata Journal 17(3) pp. 630-651.

BibTeX

Bugs

There are no known bugs. The number of people who have used the code until now is small though, so please report any unintend or surprising behaviour.

Changelog

  • 1.1.0 added an option (fixlevels()) to constrain re-randomization to certain valuse of the treatment variable. This can be used for pairwise tests in multi-treatment experiments.
  • 1.0.9 added the strict and the eps option to the helpfile and added parameter-checks so that "strict" enforces "eps(0)". h/t Katharina Nesselrode
  • 1.0.8 minor bugfix and I got rid of the google analytics part
  • 1.0.7 Jason Kerwin pointed out that when string variables were used as strata or cluster ids, all observations were treated as belonging to the same. This is fixed with this version. Also I sped up execution time by dropping unneeded code.
  • 1.0.6 Jason Kerwin pointed out an issue with the "saveresampling()"-option. This version fixes this.
  • 1.0.4 David McKenzie pointed out that under some conditions, the random seed was ignored. This is fixed with this version.
  • 1.0.3 is the version that was published in the Stata Journal.

Media Coverage

Disclaimer of Warranties and Limitation of Liability

Use at own risk. You agree that use of this software is at your own risk. The author is optimistic but does not make any warranty as to the results that may be obtained from use of this software. The author would be very happy to hear about any issues you might find and will be transparent about changes made in response to user inquiries.