PlackettLuce (version 0.4.3)

aggregate: Aggregate Rankings

Description

Aggregate rankings, returning an "aggregated_rankings" object of the unique rankings and their frequencies. The frequencies can be extracted via the function freq().

Usage

# S3 method for rankings
aggregate(x, freq = NULL, ...)
as.aggregated_rankings(x, ...)
# S3 method for aggregated_rankings
[(x, i, j, ..., drop = FALSE, as.aggregated_rankings = TRUE)
freq(x)

Value

A data frame of class "aggregated_rankings", with columns

ranking

A "rankings" object of the unique rankings.

freq

The corresponding frequencies.

Methods are available for rbind() and as.matrix().

Arguments

x

A "rankings" object for aggregate(); an object that can be coerced to a "aggregated_rankings" object for as.aggregated_rankings(), otherwise an "aggregated_rankings" object.

freq

A vector of frequencies for rankings that have been previously aggregated.

...

Additional arguments, currently unused.

i

indices specifying rankings to extract, as for [.

j

indices specifying items to extract, as for [.

drop

if TRUE return single row/column matrices as a vector.

as.aggregated_rankings

if TRUE create an "aggregated_rankings" object from the indexed rankings. Otherwise index the underlying matrix of ranks and return in a data frame with the corresponding frequencies.

See Also

preflib() for an object that can be coerced to an "aggregated_rankings" object.

Examples

# create a rankings object with duplicated rankings
R <- matrix(c(1, 2, 0, 0,
              0, 1, 2, 3,
              2, 1, 1, 0,
              1, 2, 0, 0,
              2, 1, 1, 0,
              1, 0, 3, 2), nrow = 6, byrow = TRUE)
colnames(R) <- c("apple", "banana", "orange", "pear")
R <- as.rankings(R)

# aggregate the rankings
A <- aggregate(R)

# subsetting applies to the rankings, e.g. first two unique rankings
A[1:2]

# (partial) rankings of items 2 to 4 only
A[, 2:4]

# convert to a matrix
as.matrix(A)

# frequencies are automatically used as weights by PlackettLuce()
mod <- PlackettLuce(A)
mod$weights