More statistics to consider #3
Description
While I think there is already enough for most needs, we can always add more if they are straightforward to calculate. If so, perhaps this should be an optional argument with default to FALSE.
- Yule/Phi (will not call it Matthews!)
- Tetrachoric correlation (probably not due to heavy lifting)
- AUC (see psych implementation)
- d'
- Peirce's coefficient (because he was awesome)
- Jaccard (still notably used in some disciplines)
- Odds ratio
- Others
Possibly separate correlation coefficients
Preference for those that are zero if statistically independent
See this historical overview for a ridiculous number of things to calculate: A Survey of Binary Similarity and Distance Measures Seung-Seok Choi, Sung-Hyuk Cha, Charles C. Tappert
Also:
Warrens, Matthijs (2008), On Association Coefficients for 2x2 Tables and Properties That Do Not Depend on the Marginal Distributions.