Binary comparisons from a ranking object. Ties are not taken into account, then they are added as NA's.
set_binomialfreq(object, drop.null = FALSE, disaggregate = FALSE)A data.frame with binary rank of pairwise contests:
a factor with n levels for the first player in the contests
a factor with n levels (same as player1) for the second player in the contests
number of times player1 wins against player2
number of times player2 wins against player1
Turner H. & Firth D. (2012). Journal of Statistical Software, 48(9), 1–21. doi:10.18637/jss.v048.i09
Other rank functions:
rank_numeric(),
rank_tricot(),
rank_tricot2(),
set_paircomp()
library("PlackettLuce")
R = matrix(c(1, 2, 0, 0,
4, 1, 2, 3,
2, 4, 3, 1,
1, 2, 3, 0,
2, 1, 1, 0,
1, 0, 3, 2), nrow = 6, byrow = TRUE)
colnames(R) = c("apple", "banana", "orange", "pear")
R = as.rankings(R)
set_binomialfreq(R)
#> player1 player2 win1 win2
#> <fct> <fct> <int> <int>
#> 1: apple banana 3 2
#> 2: apple orange 3 2
#> 3: banana orange 2 1
#> 4: apple pear 1 2
#> 5: banana pear 1 1
#> 6: orange pear 1 2
set_binomialfreq(R, disaggregate = TRUE)
#> id player1 player2 win1 win2
#> <int> <fct> <fct> <int> <int>
#> 1: 1 apple banana 1 0
#> 2: 2 apple banana 0 1
#> 3: 2 apple orange 0 1
#> 4: 2 banana orange 1 0
#> 5: 2 apple pear 0 1
#> ---
#> 17: 5 apple banana 0 1
#> 18: 5 apple orange 0 1
#> 21: 6 apple orange 1 0
#> 41: 6 apple pear 1 0
#> 61: 6 orange pear 0 1