Mann-Kendall Tau Test with Knight's Algorithm — TH_MK_Corr • TimeHiVE

Efficient implementation of the Mann-Kendall test for correlation using Knight's algorithm of (O(n log n) complexity). Handles ties correctly.

Usage

TH_MK_Corr(x, y, alternative = c("two.sided", "greater", "less"))

Arguments

x: a numerical vector
y: a numerical vector
alternative: Type of alternative hypothesis ("two.sided", "greater", "less")

Value

A htest object with results

statistic: Kendall's tau-b statistic
p.value: p-value
alternative: Alternative hypothesis description

Author

Vladimiro Andrea Boselli, (2025) boselli.v@irea.cnr.it

Examples

if (FALSE) { # \dontrun{
kt <- TH_MK_Corr(sub_series1, sub_series2, alternative = "less")
c(kt$statistic, kt$p.value)
} # }