I am looking for an equivalent R-package for 'rwolf' in Stata for multiple hypothesis correction. So far I have found boot_stepdown in the multitestr package which is based on Westfall & Young and should be equivalent to wyoung in Stata. However, I am looking for a procedure based on Romano-Wolf. Does anyone know a suitable R-package?
Thanks!!
A quick search gives me this: https://cran.r-project.org/web/packages/StepwiseTest/
It does look like what you're looking for? Also, a more thorough search gives possible results in packages hdm and NRejections.
Note that some packages cite Romano and Wolf 2007, others Romano and Wolf 2005, I don't know if it means two different algorithms or just one paper more theoretical the other more practical.
Thank you very much for your reply. I looked into the corr_tests in the NRejections package now. Here it is possible to specify the method, one option is "romano". However, when I specify "romano" the following error is returned:
error: Mat::row(): index out of bounds
All other methods (e.g. nreject, Wstep, holm) work out fine. I have already checked, there is no NAN in my dataset.
Do you have any idea what the problem could be?
No, sorry, I don't have experience with any of these packages.
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.
If you have a query related to it or one of the replies, start a new topic and refer back with a link.