The Bertin Classification Criterion

Computes the Bertin Classification Criterion for a contingency table of any dimensions.

BCC(x)

Arguments

x

A data matrix, table or array.

Details

The BCC counts the number of observation pairs which differ in all variables but are not fully concordant, (i.e. neither of the two observations of each pair is larger than the other in all variables).

Value

The criterion value.

See also

kendalls

Examples

M <-arsim(1000, c(12,12), 3)
BCC(M)
#> [1] 180555

M2 <- optile(M, iter = 100)
BCC(M2)
#> [1] 41720