Distance Based Ranking Models

Implements distance based probability models for ranking data. The supported distance metrics include Kendall distance, Spearman distance, Footrule distance, Hamming distance, , Weighted-tau distance and Weighted Kendall distance. Phi-component model and mixture models are also supported.


R package rankdist

Implements distance based probability models for ranking data. The supported models include Mallow's Phi-model (Kendall distance), Weighted Kendall distance model and Phi-Component model. Mixture models are also supported.

Examples

Fit a single-cluster Mallow's Phi-model to APA election data.

library(rankdist)
testctrl <- new("RankControlKendall")
testinit <- new("RankInit",param.init=list(1),modal_ranking.init=list(c(2,3,4,1,5)),clu=1L)
ken_c1 <- RankDistanceModel(apa_obj,testinit,testctrl)

Fit a three-cluster Weighted Kendall distance model to APA election data.

testinit <- new("RankInit",param.init=list(rep(0.1,4),rep(0.1,4),rep(0.1,4)),modal_ranking.init=list(c(3,4,5,1,2),c(2,3,1,5,4),c(4,2,5,3,1)),clu=3L,p.init=rep(1,3)/3)
testctrl <- new("RankControlWeightedKendall") ```

News

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("rankdist")

1.1.3 by Zhaozhi Qian, 9 months ago


Browse source code at https://github.com/cran/rankdist


Authors: Zhaozhi Qian


Documentation:   PDF Manual  


GPL (>= 2) license


Imports Rcpp, hash, optimx, permute, methods

Suggests testthat

Linking to Rcpp


See at CRAN