kendallknight: Efficient Implementation of Kendall's Correlation Coefficient Computation

The computational complexity of the implemented algorithm is \(O(n \log(n))\), which is faster than the base R implementation in cor with method = "kendall" that has a computational complexity of \(O(n^2)\). For small vectors (i.e., less than 100 observations), the time difference is negligible. However, for larger vectors, the difference can be substantial. The references are Knight (1966) doi:10.2307/2282833 , Abrevaya (1999) doi:10.1016/S0165-1765(98)00255-9 , Christensen (2005) doi:10.1007/BF02736122 and Emara (2024) https://learningcpp.org/. This implementation is described in Vargas Sepulveda (2024) doi:10.48550/arXiv.2408.09618 .

See also

Useful links:

• https://pacha.dev/capybara/

• https://github.com/pachadotdev/capybara

• Report bugs at https://github.com/pachadotdev/capybara/issues

Author

Maintainer: Mauricio Vargas Sepulveda m.sepulveda@mail.utoronto.ca (ORCID)

Other contributors:

• Loader Catherine (original stirlerr implementations in C (2000)) [contributor]

• Ross Ihaka (original chebyshev_eval, gammafn and lgammacor implementations in C (1998)) [contributor]