PlackettLuce: plfit – R documentation

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

plfit

PlackettLuce Wrapper for Model-based Recursive Partitioning

Description

This is a wrapper around PlackettLuce as required by mob for model-based recursive partitioning. It is not intended for general use.

Usage

plfit(
  y,
  x = NULL,
  ref = 1L,
  start = NULL,
  weights = NULL,
  offset = NULL,
  ...,
  estfun = FALSE,
  object = FALSE
)

Arguments

`y`	a `"grouped_rankings"` object giving the rankings to model.
`x`	unused.
`ref`	An integer or character string specifying the reference item (for which log worth will be set to zero). If `NULL` the sum of the log worth parameters is set to zero.
`start`	starting values for the worth parameters and the tie parameters on the raw scale (worth parameters need not be scaled to sum to 1). If `normal` is specified, `exp(normal$mu)` is used as starting values for the worth parameters. Coefficients from a previous fit can be passed as the result of a call to `coef.PlackettLuce`, or the `coefficients` element of a `"PlackettLuce"` object.
`weights`	an optional vector of weights for each ranking.
`offset`	unused.
`...`	additional arguments passed to `PlackettLuce`.
`estfun`	logical. If `TRUE` the empirical estimating functions (score/gradient contributions) are returned.
`object`	logical. If `TRUE` the fitted model is returned.

Value

a list with elements

`coefficients`	model coefficients.
`objfun`	the negative log-likelihood.
`estfun`	if `estfun` the empirical estimating functions.
`object`	if `object` the fitted model.

Examples

# rankings
R <- matrix(c(1, 2, 0, 0,
              4, 1, 2, 3,
              2, 1, 1, 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)

# group rankings into two groups
G <- group(R, rep(1:2, 3))

# plfit() gives the same results as PlackettLuce()
pl <- plfit(G)
pl$coefficients
-pl$objfun

mod <- PlackettLuce(R)
coef(mod)
logLik(mod)

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