Releases: hturner/PlackettLuce
Releases · hturner/PlackettLuce
v0.3.2
v0.3.1
v0.2-9
vcov.PlackettLuce()
works again forref = NULL
(bug introduced with vcov method in version 0.2-4)- avoid dependency on R >= 3.6.0 by providing alternatives to
asplit()
read.soi()
andread.toi()
now handle incomplete rankings with very irregular lengths correctly.read.*()
functions for Preflib formats now give a meaningful error when the file or URL does not exist, and a warning if the file is corrupt.as.rankings
withinput = "orderings"
now checks coded values can be matched to item names, if provided.PlackettLuce()
now works withnspeudo > 0
when there are no observed paired comparisons.?PlackettLuce
now gives advice on analysing data with higher order ties.
v0.2-8
New Features
- New
"aggregated_rankings"
object to store aggregated rankings with the corresponding frequencies. Objects of class"rankings"
can be aggregated via theaggregate
method; alternativelyrankings()
andas.rankings()
will create an"aggregated_rankings"
object whenaggregate = TRUE
.as.rankings()
also handles pre-aggregated data, accepting frequencies via thefreq
argument. - New
freq()
function to extract frequencies from aggregated rankings. as.rankings()
can now create a"grouped_rankings"
object, if a grouping index is passed via theindex
argument.- New
as.matrix()
methods for rankings and aggregated rankings to extract the underlying matrix of rankings, with frequencies in the final column if relevant. This means rankings can be saved easily withwrite.table()
. - New
complete()
anddecode()
functions to help pre-process orderings before converting to rankings,complete()
infers the item(s) in r'th rank given the items in the other (r - 1) ranks.decode()
converts coded (partial) orderings to orderings of the items in each ordering. - New
read.soi()
,read.toc()
andread.toi()
to read the corresponding PrefLib file formats (for data types "Strict Orders - Incomplete List", "Orders with Ties - Complete List" and "Orders with Ties - Incomplete List"). Anas.aggregated_rankings()
method is provided to convert the data frame of aggregated orderings to an"aggregated_rankings"
object.
Improvements
pltree()
now respectsna.action
and will pad predictions and fitted values forna.action = "na.exclude"
if the rankings are missing for a whole group or one of the model covariates has a missing value.PlackettLuce()
now has anna.action
argument for handling of missing rankings.fitted()
andchoices()
now return data frames, with list columns as necessary.
Changes in behaviour
rankings()
now sets redundant/inconsistent ranks toNA
rather than dropping them. This does not affect the final ranking, unless it is completelyNA
.- The frequencies column in the data frame returned by
read.soc()
is now namedFreq
rather thann
. - The
"item"
attribute of the data frame returned byread.soc()
is now named"items"
. - The
labels
argument inas.rankings()
has been deprecated and replaced byitems
. grouped_ranking()
has been deprecated and replaced bygroup()
.- The redundant columns in the
nascar
data have been dropped.
v0.2-6
v0.2-5
New Features
PlackettLuce()
now supports MAP estimation with a multivariate normal prior on log-worths and/or a gamma prior on ranker adherence.PlackettLuce()
now returns log-likelihood and degrees of freedom for the null model (where all outcomes, including ties, have equal probability).- There is now a
vcov
method for Plackett-Luce trees.
Changes in Behaviour
itempar.PlackettLuce()
now always returns a matrix, even for a single node tree.
Bug Fixes
pltree()
orPlackettLuce()
with grouped rankings now work correctly with weights.
v0.2-3
Improvements
- Print methods for
"PlackettLuce"
and"summary.PlacketLuce"
objects now respectoptions("width")
.
Changes in Behaviour
fitted
always returnsn
which is now weighted count of rankings (previously only returned unweighted count with argumentaggregate = TRUE
).
Bug fixes
- Correct vcov for weighted rankings of more than two items.
- Enable
AIC.pltree
to work on"pltree"
object with one node.
v0.2-2
New features
- Add
AIC.pltree
to enable computation of AIC on new observations (e.g.
data held out in cross-validation). - Add
fitted.pltree
to return combined fitted probabilities for each choice within each ranking, for each node in a Plackett-Luce tree.
Bug fixes
vcov.PlackettLuce
now works for models with non-integer weights (fixes #25).plot.pltree
now works forworth = TRUE
with psychotree version 0.15-2 (currently pre-release on https://r-forge.r-project.org/R/?group_id=330)PlackettLuce
andplfit
now work whenstart
argument is set.itempar.PlackettLuce
now works withalias = FALSE
v0.2-1
New features
- Add pkgdown site.
- Add content to README (fixes #5).
- Add
plot.PlackettLuce
method so that plotting works for a saved"PlackettLuce"
object
Improvements
- Improved vignette, particularly example based on
beans
data (which has been updated). - Improved help files particularly
?PlackettLuce
and newpackage?PlackettLuce
. (Fixes #14 and #21).
Changes in behaviour
maxit
defaults to 500 inPlackettLuce
.- Steffensen acceleration only applied in iterations where it will increase the log-likelihood (still only attempted once iterations have reached a solution that is "close enough" as specified by
steffensen
argument).
Bug fixes
coef.pltree()
now respectslog = TRUE
argument (fixes #19).- Fix bug causes lack of convergence with iterative scaling plus pseudo-rankings.
[.grouped_rankings]
now works for replicated indices.
v0.2-0
New Features
- Add vignette.
- Add data sets
pudding
,nascar
andbeans
. - Add
pltree()
function for use withpartykit::mob()
. Requires new objects of type"grouped_rankings"
that add a grouping index to a"rankings"
object and store other derived objects used byPlackettLuce
. Methods to print, plot and predict Plackett-Luce trees are provided. - Add
connectivity()
function to check connectivity of a network given adjacency matrix. Newadjacency()
function computes adjacency matrix without creating edgelist, so removeas.edgelist
generic and method for `"PlackettLuce" objects. - Add
as.data.frame
methods so that rankings and grouped rankings can be added to model frames. - Add
format
methods for rankings and grouped_rankings, for pretty printing. - Add
[
methods for rankings and grouped_rankings, to create valid rankings from selected rankings and/or items. - Add method argument to offer choices of iterative scaling (default), or direct maximisation of the likelihood via BFGS or L-BFGS.
- Add
itempar
method for "PlackettLuce" objects to obtain different parameterizations of the worth parameters. - Add
read.soc
function to read Strict Orders - Complete List (.soc) files from http://www.preflib.org.
Changes in behaviour
Old behaviour should be reproducible with arguments
npseudo = 0, steffensen = 0, start = c(rep(1/N, N), rep(0.1, D))
where N
is number of items and D
is maximum order of ties.
- Implement pseudo-data approach - now used by default.
- Improve starting values for ability parameters
- Add Steffensen acceleration to iterative scaling algorithm
- Dropped
ref
argument fromPlackettLuce
; should be specified instead when callingcoef
,summary
,vcov
oritempar
. qvcalc
generic now imported from qvcalc
Improvements
- Refactor code to speed up model fitting and computation of fitted values and vcov.
- Implement ranking weights and starting values in
PlackettLuce
. - Add package tests
- Add
log
argument tocoef
so that worth parameters (probability of coming first in strict ranking of all items) can be obtained easily.