Moore-rayleigh

[J-Wall]

This PR implements the Moore's modified Rayleigh test described in Moore, B. R. (1980). A modification of the Rayleigh test for vector data. Biometrika, 67(1), 175-180. Addresses part of #214. I've tried to follow your style for code and documentation, and I've included tests based on R* values obtained from the Oriana circular statistics program, and manually calculated p-values.

This test is slightly obscure (in fact, I couldn't find it in any other statistics library...) although it's used a fair bit in animal migration research. The lack of an open-source implementation was bugging me, so I thought I would contribute to an existing library rather than start my own. Cheers!

[raphaelvallat]

Hi @J-Wall!

Thanks so much for this. Really clean PR! Two comments after a quick review:

1. For unit testing, can you get higher float precision than 3 decimals? I'm not familiar at all with the Oriana software.

2. Should we combine the circ_moore_rayleigh and circ_rayleigh into a single function? e.g. the Moore's method could be specified as an optional parameter to the main circ_rayleigh function? I'd much prefer this option rather than burdening Pingouin's API with a very rare function.

Thanks,
Raphael

[codecov]

Codecov Report

Merging #215 (5475d88) into master (cc5d687) will decrease coverage by 0.08%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #215      +/-   ##
==========================================
- Coverage   99.02%   98.93%   -0.09%     
==========================================
  Files          19       19              
  Lines        3280     3292      +12     
  Branches      529      529              
==========================================
+ Hits         3248     3257       +9     
- Misses         16       18       +2     
- Partials       16       17       +1     

Impacted Files	Coverage Δ
pingouin/circular.py	100.00% <100.00%> (ø)
pingouin/plotting.py	96.48% <0.00%> (-0.88%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update cc5d687...5475d88. Read the comment docs.

[raphaelvallat]

I am closing this PR since there has not been any activity in several months. Feel free to reopen. Thanks