compare_ranks: Count occurrences of pairs in rankings

In netrankr: Analyzing Partial Rankings in Networks

Count occurrences of pairs in rankings

Description

Counts the number of concordant, discordant and (left/right) ties between two rankings.

Usage

compare_ranks(x, y)

Arguments

x	A numeric vector.
y	A numeric vector with the same length as x.

Details

Explicitly calculating the number of occurring cases is more robust than using correlation indices as given in the cor function. Especially left and right ties can significantly alter correlations.

Value

A list containing

concordant	number of concordant pairs: x[i] > x[j] and y[i] > y[j]
discordant	number of discordant pairs: x[i] > x[j] and y[i] < y[j]
ties	number of tied pairs: x[i] == x[j] and y[i] == y[j]
left	number of left ties: x[i] == x[j] and y[i] != y[j]
right	number of right ties: x[i] != x[j] and y[i] == y[j]

Author(s)

David Schoch

Examples

library(igraph)
tg <- threshold_graph(100, 0.2)
compare_ranks(degree(tg), closeness(tg)) # only concordant pairs
compare_ranks(degree(tg), betweenness(tg)) # no discordant pairs
## Rank Correlation
cor(degree(tg), closeness(tg), method = "kendall") # 1
cor(degree(tg), betweenness(tg), method = "kendall") # not 1, although no discordant pairs

netrankr documentation built on Aug. 20, 2023, 5:06 p.m.