PlackettLuce: group – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

group

Group Rankings

Description

Create an object of class "grouped_rankings" which associates a group index with an object of class "rankings". This allows the rankings to be linked to covariates with group-specific values as the basis for model-based recursive partitioning, see pltree.

Usage

group(x, index, ...)

as.grouped_rankings(x, ...)

## S3 method for class 'paircomp'
as.grouped_rankings(x, ...)

## S3 method for class 'grouped_rankings'
x[i, j, ..., drop = TRUE, as.grouped_rankings = TRUE]

## S3 method for class 'grouped_rankings'
format(x, max = 2L, width = 20L, ...)

Arguments

`x`	a `"rankings"` object for `group()`; an object that can be coerced to a `"grouped_rankings"` object for `as.grouped_rankings()`, otherwise a `"grouped_rankings"` object.
`index`	a numeric vector of length equal to the number of rankings specifying the subject for each ranking.
`...`	additional arguments passed on to `as.rankings` by `grouped_rankings` or `as.grouped_rankings`; unused by `format`.
`i`	indices specifying groups to extract, may be any data type accepted by `[`.
`j`	indices specifying items to extract, as for `[`.
`drop`	if `TRUE` return single row/column matrices as a vector.
`as.grouped_rankings`	if `TRUE` return a grouped_rankings object, otherwise return a matrix/vector.
`max`	the maximum number of rankings to format per subject.
`width`	the maximum width in number of characters to format each ranking.

Value

An object of class "grouped_rankings", which is a vector of of group IDs with the following attributes:

`rankings`	The `"rankings"` object.
`index`	An index match each ranking to each group ID.
`R`	A matrix with items ordered from last to first place, for each ranking.
`S`	The rankings matrix with the ranks replaced by the size of the chosen set for free choices and zero for forced choices.
`id`	A list with elements of the adjacency matrix that are incremented by each ranking.

Examples

# ungrouped rankings (5 rankings, 4 items)
R <- as.rankings(matrix(c(1, 2, 0, 0,
                          0, 2, 1, 0,
                          0, 0, 1, 2,
                          2, 1, 0, 0,
                          0, 1, 2, 3), ncol = 4, byrow = TRUE))
length(R)
R

# group rankings (first three in group 1, next two in group 2)
G <- group(R, c(1, 1, 1, 2, 2))
length(G)

## by default up to 2 rankings are shown per group, "..." indicates if
## there are further rankings
G
print(G, max = 1)

## select rankings from group 1
G[1,]

## exclude item 3 from ranking
G[, -3]

## rankings from group 2, excluding item 3
## - note group 2 becomes the first group
G[2, -3]

## index underlying rankings without creating new grouped_rankings object
G[2, -3, as.grouped_rankings = FALSE]

PlackettLuce

Plackett-Luce Models for Rankings

v0.4.0

GPL-3

Authors

Heather Turner [aut, cre] (<https://orcid.org/0000-0002-1256-3375>), Ioannis Kosmidis [aut] (<https://orcid.org/0000-0003-1556-0302>), David Firth [aut] (<https://orcid.org/0000-0003-0302-2312>), Jacob van Etten [ctb] (<https://orcid.org/0000-0001-7554-2558>)

Initial release