Plackett-Luce rankings from numeric values

Group and coerce numeric values into Plackett-Luce rankings.

rank_numeric(
  data,
  items,
  input,
  id = NULL,
  group = FALSE,
  ascending = FALSE,
  ...
)

Arguments

data

a data.frame with columns specified by items and input values

items

a character or numerical vector for indexing the column(s) containing the item names in data

input

a character or numerical vector for indexing the column(s) containing the values in data to be ranked

id

an index of data indicating the ids for "long" data

group

logical, if TRUE return an object of class "grouped_rankings"

ascending

logical, only for floating point numbers, to compute rankings from lower to higher values

...

additional arguments passed to methods

Value

a PlackettLuce "rankings" object, which is a matrix of dense rankings

See also

rankings

Other rank functions: rank_tricot(), set_binomialfreq(), set_paircomp()

Author

Kauê de Sousa

Examples

# A matrix with 10 rankings of 5 items (A, B, C, D, E)
# with numeric values as "rank"
set.seed(123)
df = cbind(id = rep(1:10, each = 5),
            items = rep(LETTERS[1:5], times = 10),
            input = runif(50, 1, 3))

# return an object of class 'rankings'
R = rank_numeric(df,
                  items = 2,
                  input = 3,
                  id = 1)

# rankings can be computed in ascending order
R = rank_numeric(df,
                  items = 2,
                  input = 3,
                  id = 1,
                  ascending = TRUE)


# return an object of class 'grouped_rankings'
R = rank_numeric(df,
                  items = 2,
                  input = 3,
                  id = 1,
                  group = TRUE)