1

I would like to compare two samples with the watson's two sample test in R (circular data). As I have a lot of ties in my samples, I followed the example given in the book from Zar (Biostatistical analysis) for the watson's two sample test with ties (I did it by hand). I was curious to see if it would give me the same result with the code developped in R (watson.two.test). I saw some differences so I guess the test available in R is not adapted when there are a lot of ties. May I ask you if it exists on R a code for this test with ties?

Thanks a lot for your answer.

nathmar
  • 31

1 Answers1

0

I am not sure if it could help someone but in any case I think I found a way to solve my problem. I just read a paper from Jammalamadaka et al. 2021 (https://link.springer.com/article/10.1007/s13571-020-00244-9) where they describe circular tests for low sample size using spacing frequencies instead of ranking. This solve as well the problems with the ties. I advise to read if you are working in these type of data. The Rcode is taken from here: https://github.com/SMAC-Group/TwoCircles#readme

Hope it will help someone else!

nathmar
  • 31